Unraveling city-specific signature and identifying sample origin locations for the data from CAMDA MetaSUB challenge

Unraveling city-specific signature and identifying sample origin locations for the data from CAMDA MetaSUB challenge | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Unraveling city-specific signature and identifying sample origin locations for the data from CAMDA MetaSUB challenge Runzhi Zhang, Alejandro R. Walker, Susmita Datta This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.2.20675/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 04 Jan, 2021 Read the published version in Biology Direct → Version 1 posted 14 You are reading this latest preprint version Abstract Background Composition of microbial communities can be location specific, and the different abundance of taxon within location could help us to unravel city-specific signature and predict the sample origin locations accurately. In this study, the whole genome shotgun (WGS) metagenomics data from samples across 16 cities around the world and samples from another 8 cities were provided as the main and mystery datasets respectively as the part of the CAMDA 2019 MetaSUB “Forensic Challenge”. The feature selection, normalization, three methods of machine learning, PCoA (Principal Coordinates Analysis) and ANCOM (Analysis of composition of microbiomes) were conducted for both the main and mystery datasets. Results Feature selection, combined with the machines learning methods, revealed that the combination of the common features was effective for predicting the origin of the samples. The average error rates of 11.6% and 30.0% of three machine learning methods were obtained for main and mystery datasets respectively. Using the samples from main dataset to predict the labels of samples from mystery dataset, nearly 89.98% of the test samples could be correctly labeled as “mystery” samples. PCoA showed that nearly 60% of the total variability of the data could be explained by the first two PCoA axes. Although many cities overlapped, the separation of some cities was found in PCoA. The results of ANCOM, combined with importance score from the Random Forest, indicated that the common “family”, “order” of the main-dataset and the common “order” of the mystery dataset provided the most efficient information for prediction respectively. Conclusions The results of the classification suggested that the composition of the microbiomes was distinctive across the cities, which was also supported by the results from ANCOM and importance score from the RF. The analysis utilized in this study can be of great help in field of forensic science to efficiently predict the origin of the samples. And the accurate of the prediction could be improved by more samples and better sequencing depth. General Microbiology Microbiome OTU WGS feature selection machine learning Random Forest Support Vector Machine Linear Discriminant Analysis PCoA ANCOM Figures Figure 1 Figure 2 Figure 3 Figure 4 Background The advent of next generation sequencing (NGS) technologies for metagenomics has experienced a tremendous improvement, which allows the generation of large sequence datasets derived from diverse ecosystems, such as the human body, soil, and ocean water [1]. The use of whole genome sequencing (WGS) has been reported to have multiple advantages when compared with the 16S rRNA amplicon data [2]. As the composition of microbial communities can be location specific [3], studying the microbiome from different cities improves our understanding of city-specific microbes and their contributions to ecosystem composition and diversity. In this work, we aimed to unravel city-specific signature and find the appropriate features for identifying and predicting the origin location of samples from different areas. The dataset was provided by MetaSUB ( http://camda2019.bioinf.jku.at/doku.php/contest_dataset ), which aimed to build an international metagenomic map of urban spaces, based on extensive sampling of mass-transit system and other public areas around the world. They partnered with CAMDA for an early release of microbiome data obtained from global City Sampling Days, comprising the WGS metagenomics data. The main dataset covered 16 cities across the globe, with tens of samples per city. Moreover, one more dataset with 8 cities was provided as mystery set from the CAMDA 2019 MetaSUB challenge to serve as testing samples. And the true city-information of the mystery data was provided much later in the process. Table 1 presented a tabulated insight of the data for all the cities. Results Feature selection First, we selected the common “species”, “family”, and “order” existing across all the 16 cities and the number of variables was 7, 9 and 9 respectively. The number of features was limited and more information about the microbes was needed. In order to increase the number of features, we selected the features based on additional rules: a) The features were selected based on the ubiquity of the “species”, “family” and “order” across all the cities. b) The features were selected based on the ubiquity of the “species”, “family” and “order” across all the samples. c) The combinations of the common “species”, “family”, and “order” were regarded as the combined features. Table 2 presented the details of the features selected based on additional rules. For simplicity, the mystery dataset was analyzed based on the common features and combined features. After feature selection, the aggregated raw counts of each dataset were normalized to generate log2-cpm for further analyses. Machine learning analysis For the main and mystery datasets respectively, results from Random forest (RF) [4], Support Vector Machine (SVM) [5] and Linear Discriminant Analysis (LDA) [6] were obtained with the leave-one-out cross validation (CV), one test sample was randomly selected in each run with 1000 runs repeated. Table 3 presented the details of the classification error rate based on different rules using the main and mystery datasets respectively. As seen in the table, when the features existing in at least N cities were selected for the analysis, the changing trends of the error rate of RF-species (qualitied “species” was used for analysis using RF), SVM-species, LDA-species, LDA-family and LDA-order shared a similar pattern, a decreased CV error rate was obtained when increasing the number of features, and then the lowest error rate was achieved. For RF-family and SVM-family, the error rate hasn’t changed considerably when we increased the number of the variables at the family rank. Therefore, using the common “family” was better, as we obtained the low error rate without including too many features. And for RF-order and SVM-order, the lowest error rate was obtained using common “order”. Additionally, when top M features with the highest ubiquity across all the samples were selected for analysis, error rates decreased with the increasing number of features used and then the lowest error rate was achieved, no matter which machine learning methods or which kinds of feature we used for analysis. In addition, for the combined features, the best performance was achieved using the features combined with common “species”, “family” and “order” (7 species, 9 families, 9 orders), the error rates obtained from RF, SVM and LDA were 11.6%, 11.5% and 11.8% respectively. And for the mystery dataset, we obtained the lowest average error rate when we used the combined features with 8 species and 15 orders. The error rates were 29.7%, 32.1% and 28.2% for RF, SVM and LDA respectively. The error rates of predictions for each city was presented in Figure 1 . From Figure 1.A , the low error rates were obtained from the cities with better sequencing depth ( Table S1 ) including Bogota, Ilorin, New York, Offa, Sacramento and Tokyo. The average error rates of the three methods for these four cities were 7.84%, 5.49%, 2.03%, 6.31%, 3.08% and 5.80% respectively. However, the error rates of Auckland and Hamilton, two cities also with good sequencing depth, were 26.39% and 18.44% respectively. By looking into the details of the results, we found that the samples AKL_1, AKL_7, AKL_14 and HAM_7, HAM_12 cannot be predicted correctly by all the three methods. In other words, the microbial composition of these samples was different from the other samples in Auckland and Hamilton ( Table S2 ), making them difficult to be identified. And we found that the samples from London with the poor sequencing depth showed low error rate. Upon finding excessive zeros in samples from London, the samples of London could be easily identified from all the samples, which resulted in the low error rate of London. In addition, a possible explanation for low error rate could be the insufficient number of samples, as the cities with the lowest number of samples including Sofia and Marseille showed high error rates. And from Figure 1.B , the cities with deep sequencing ( Table S3 ), i.e. Oslo and Rio de Janeiro, were the two of the first three cities with the lowest average error rate (6.64% and 16.81% respectively). Similarly, the cities with poor sequencing depth such as Doha and Brisbane, showed the high average error rate (85.71% and 62.61% respectively). And Doha, also with the limited number of samples, achieved the highest error rate among all the cities. The evident from the Figure 1 that most cities with limited samples and poor sequencing depth had high error rates, indicating that sufficient samples and deep sequencing were necessary for successfully predicting the provenance of samples. In addition to the analyses based on the main and mystery datasets respectively, we have also used the prediction models built based on the main dataset to predict the samples from mystery dataset. As the main dataset and mystery dataset had no cities in common, the information of cities from mystery dataset was lacked in the main dataset. Therefore, 50% of the samples of each city from mystery dataset were randomly sampled and added to the main dataset to serve as the part of the training dataset, and the rest of the samples from mystery dataset were served as the test samples. In the training dataset, the samples from mystery dataset were labeled as the “mystery”. The prediction models were built based on the common “family” (5 families) and “order” (6 orders), as there was no common “species” between the main and mystery datasets. Three different classifications were used and the random samplings were conducted for 1000 times. For each run, the predicted labels were checked with the real labels. The average error rates for RF, SVM and LDA were 10.48%, 9.21% and 10.36% respectively, indicating that the prediction models could effectively identify the mystery samples from all the samples. The following analyses were based on the features which achieved the lowest average error rate. Principal Coordinates Analysis The results of PCoA [7] in Figure 2 presented the bi-plots for both datasets. Figure 2.A illustrated the main dataset and the 58.4% of total variability of the data could be explained by the first two PCoA axes. A separation of the cities could be referred from the plot. For example, London was separated from most cities and on the rightmost site, which was corresponding to the result from machine learning methods, indicating that the excessive zeros in samples from London made the prediction easier. However, many cities overlapped together. Specifically, Ilorin and Offa, both the cities of Nigeria, whose ellipses showed a massive overlap. Also, Auckland and Hamilton, both being in New Zealand, overlapped with each other. The result of mystery dataset was given in Figure 2.B . The first two PCoA axes explained 65.4% of total variability in the data, which was comparable with the percentage explained in the main dataset. Although many cities overlapped, samples of Oslo were clustered together and distributed at the top of the plot, separating from the most samples. Also, some of the samples from Rio de Janeiro were far away from the most cities, which made these samples easier to be identified. These results were corresponding to the low error rates of Oslo and Rio de Janeiro in the previous section. Analysis of composition of microbiomes The results from the analysis of composition of microbiomes (ANCOM) [8] were presented in Figure 3 . The relative abundances of the features were used to conduct the pair-wise comparisons among all the cities. Upon the significance of the features, the differentially abundant features were found. The features, which served as the predictors, on the right were ordered by the number of times the relative abundance was significantly different in the pair-wise comparisons. As presented in Figure 3.A , the top 10 features, i.e. Bacillaceae , Bacillales , Actinomycetales , Sphingomonadaceae , Pseudomonadaceae , Pseudomonas.spp , Sphingomonadales , Lactobacillales , streptococcaceae and Enterobacteriaceae showed the highest counts. For the top feature, the count of Bacillaceae is 43 out of 120, which meant that in the 120 pair-wise comparisons among the 16 cities, Bacillaceae was found to be significantly different in 43 comparisons. Similarly, in Figure 3.B , the top 10 features were Bacillales , Clostridiales , Pseudomonadales , Staphylococcus.epidermidis , Lactobacillales , Rhodospirillales , Flavobacteriales , Streptophyta , Burkholderiales and Enterobacteriales . Additionally, little change was seen from the importance score ( Figure 4 ) derived from the RF. It could be inferred from Figure 4.A that Bacillales, Actinomycetales , Sphingomonadales , Sphingomonadaceae , Enterobacteriaceae and streptococcaceae were also in top 10 features. Also, for the mystery dataset, Bacillales , Streptophyta, Rhodospirillales , Enterobacteriales , Lactobacillales , Clostridiales and Pseudomonadales were in the top 10 of both lists. In summary, for the main dataset, the common “family” and “order” of the OTU provided the most informative data for predicting the origins of the samples, which was also corresponding to the low error rate of three methods when using the common “family” and “order” only compared with using the common “species”. For the mystery dataset, the results of the ANCOM and feature importance combined with the error rates from three methods indicated that the common “order” dominated the prediction. Discussion and conclusions For the CAMDA challenge MetaSUB data of this year, 16 cities were included and 10 or more samples were collected for each city in the main dataset. The feature selection, normalization, three methods of machine learning algorithms, PCoA and ANCOM were conducted for both the main and mystery datasets. Combining the feature selection and the results of machine learning methods, we found that the machine learning analysis was effective in predicting the origin of the samples when the combined common features were used, indicating that the combination of the common features could be the effective microbial fingerprint for unraveling city-specific signature and identifying sample origin locations, which proved to be with great potential for forensic science. Therefore, more taxonomic ranks of microbiomes such as class and genus could be added to the combination to investigate the performance of the prediction in the future works. As in this study, by combining the common features, we obtained the low error rate of prediction without including too many features for both datasets. Additionally, we have used the main dataset and half of mystery samples as our training dataset to predict the labels of another half of mystery samples. As the error rates were 10.48%, 9.21% and 10.36% respectively for RF, SVM and LDA, most of the samples from mystery dataset could be identified correctly by the classification. PCoA analysis for both datasets showed that nearly 60% of the total variability of the data could be explained by the first two PCoA axes, most cities overlapped with each other. However, some cities such as London and Oslo were separated from most cities, indicating the unique composition of microbiomes in these cities, which was also validated by ANCOM analysis and was corresponding to the low error rates in machine learning analysis. The heatmaps of ANCOM revealed that some of the features, such as some of the common “family” and “order” in the main dataset and the common “order” in mystery dataset, were significantly different in pair-wise comparisons, and these features were also given high importance score in RF, indicating the effectiveness of these features during the classification. Additionally, this was also supported by the results of the machine learning analysis, we obtained a lower error rate using common “family” and “order” and using common “order” only for the main and mystery datasets respectively. However, due to the limited resources such as the insufficient number of samples from some cities and the poor sequencing depth of the samples, the error rate of Sofia was higher compared with other cities. Therefore, sufficient samples and better sequencing depth were necessary in order to improve the accuracy of the prediction. A similar problem was found in the mystery dataset, the number of samples was limited for most of the cities in the mystery dataset, which resulted in an overall higher error rate in the mystery dataset compared with the main dataset. And the poor sequencing of Brisbane and Doha also resulted in higher error rates compared to other cities in the mystery dataset. Additionally, the microbial composition could vary across samples within the same city, which also limited the ability of classification to correctly predict the origins of the samples. For this study, only the combined common features were selected for further analysis, some city-specific microbiomes might be ignored in this study, as they only existed in a specific city. These city-specific microbiomes could be combined with the common features to provide more powerful prediction in the future work. Methods The design of the analysis was motivated by the experience from the CAMDA 2017 and CAMDA 2018 MetaSUB Challenges [9, 10]. Compared to date from previous MeteSUB challenges, the data this year was with higher quality and deeper sequencing depth. As we have more cities included this year, the common features shared by all the cities were further limited. Therefore, the features selection was implemented this year to help us obtain the appropriate features for prediction. Finally, unsupervised and supervised techniques were used for the analyses. A more detailed description of the implementations was supplemented in the following sections. Bioinformatics and data preparation The Bioinformatics and data preparation was based on our previous paper [10]. Samples were Illumina-sequenced at different depths and delivered as FASTQ format for further analysis. Subsequent bioinformatics processing and data preparation were conducted in the “HyperGator2” high-performance cluster at the University of Florida. After quality control, the data was picked for OTUs in open-reference mode with QIIME [11]. After quality control, samples with the poor sequencing depth were removed. OTUs were aggregated as counts, and all further data processing and analysis in this study were conducted in R [12]. Feature selection Feature selection was conducted based on different rules. For the main dataset, the common “species”, “family” and “order” existing across all the 16 cities were selected at first. And three additional rules were conducted for feature selection: a) “species”, “family” and “order” existing in at least N cities were selected. N was set to 15, 14, 13, 12, 11, 10, 9 and 8 respectively, the count of the feature that did not exist in the city was marked as zero. b) “species”, “family” and “order” were reordered based on their ubiquity across all the samples respectively, the top M features with the highest ubiquity were selected. M was set to 10, 20, 30, 50, 100, 150 respectively. c) The combinations of the common “species”, “family”, and “order” were regarded as the combined features. The mystery dataset was analyzed based on the common features and combined features. The data were then normalized to generate log2-cpm to make the data meaningful using the function “voom" [13] in R package “limma” [14] for further analysis. Machine learning analysis Three classification algorithms were implemented at this stage for both the main and mystery datasets: Random Forest (RF) [4], Support Vector Machine (SVM) [5] and Linear Discriminant Analysis (LDA) [15]. When the analysis was conducted for the main and mystery datasets respectively, each method was implemented 1000 times with the leave-one-out cross validation based on the selected features. For each run, one sample was randomly selected as the test data with other samples served as the training set. RF was conducted using the R package “randomForest”, 1000 trees were used and the count of variables chosen at each split was equivalent to the square root of the number of features in the dataset. The model was fitted based on the training set and then used to predict the origin of the test sample. The result of each prediction was obtained. The overall error rate and the error rate for each city were calculated based on the result of the 1000 runs. The variable importance score [16] computed by the RF was also recorded. The SVM classifier was implemented in a similar manner. The two important parameters in SVM, i.e. gamma and c-value, affected the fitting of the models. Using the R function “best.svm” in package “e1071” [17], the SVM model with the appropriate parameters was obtained by testing the performance of models with different parameters. The LDA was conducted in a similar manner using the R package “MASS” [18]. When the models based on the main dataset were used to predict the labels of samples from mystery dataset. 50% of the samples of each city were randomly sampled from mystery dataset and added to the main dataset with the labels of “mystery” to serve as the part of the training dataset, the rest of the samples were served as the test samples. The sampling and the prediction were implemented for 1000 times. For each run, the predicted labels of test samples were checked with the real labels, a good prediction meant all test samples could be labeled as “mystery” by the classification. Principal Coordinates analysis Principal coordinates analysis (PCoA) [7] of normalized data was conducted using the R package “vegan” [19] and “ape” [20]. Firstly, the dissimilarity matrix was constructed based on Bray-Cruits distance. Then, the dissimilarity matrix was used for PCoA and a set of uncorrelated axes was generated to summarise the variability in the dataset. The two-dimensional plot was generated for assessing the separation of the cities. Analysis of composition of microbiomes Analysis of composition of microbiomes for the normalized data was conducted using the R package “ANCOM” [8], which accounted for the underlying structure in the microbial data and was used to comparing the composition of microbiomes in two or more populations. For the main and mystery datasets, ANCOM across all pair-wise comparisons in each dataset was implemented. The 120 and 28 pair-wise comparisons were made for the combination of all cities in the main and mystery datasets respectively. And the output of the ANCOM was a set of differentially abundant features between the two cities for each comparison at the significance level of 0.05. The heatmap was made based on the output for investigating the difference of microbial composition between different cities. Additionally, the results of ANCOM were compared with the importance score derived by the RF method. Declarations Ethics approval and consent to participate Not Applicable. Consent for publication Not Applicable. Availability of data and materials The datasets supporting the conclusions of this article can be obtained from the CAMDA 2019 website http://camda2019.bioinf.jku.at/doku.php/contest_dataset . Competing interests The authors declare that they have no competing interests. Funding Datta, S. was partially supported by NIH grant 1UL1TR000064 from the National Center for Advancing Translational Sciences. Author’s contributions SD reviewed the manuscript and provided theoretical support when required, RZ and ARW designed, run the analyses, RZ wrote the manuscript. All the authors have read and approved the final manuscript. Acknowledgements The samples were provided to the CAMDA 2019 competition by the MetaSUB Consortium. Authors’ information 1 Department of Biostatistics, University of Florida, 2004 Mowry Rd, Gainesville, FL, 32610, USA 2 Department of Oral Biology, University of Florida, 1395 Center Drive, Gainesville, FL, 32610, USA Abbreviations NGS: Next Generation Sequencing WGS: whole genome sequencing OTU: Operational Taxonomic Unit RF: Random Forest SVM: Support Vector Machine LDA: Linear Discriminant Analysis CV: Cross Validation PCoA: Principal Coordinates Analysis ANCOM: Analysis of composition of microbiomes References Simon C, Daniel R: Metagenomic analyses: past and future trends. Appl Environ Microbiol 2011, 77(4):1153-61. doi:10.1128/aem.02345-10. Ranjan R, Rani A, Metwally A, McGee HS, Perkins DL: Analysis of the microbiome: Advantages of whole genome shotgun versus 16S amplicon sequencing. Biochem Biophys Res Commun 2016, 469(4):967-77. doi:10.1016/j.bbrc.2015.12.083. Delgado-Baquerizo M, Oliverio AM, Brewer TE, Benavent-González A, Eldridge DJ, Bardgett RD, Maestre FT, Singh BK, Fierer N: A global atlas of the dominant bacteria found in soil. Science 2018, 359(6373):320-5. doi:10.1126/science.aap9516. Breiman L: Random Forests. Machine Learning 2001, 45(1):5-32. doi:10.1023/a:1010933404324. Cortes C, Vapnik V: Support-vector networks. Machine Learning 1995, 20(3):273-97. doi:10.1007/bf00994018. Balakrishnama S, Ganapathiraju A: Linear discriminant analysis-a brief tutorial. Institute for Signal and information Processing 1998, 18:1-8. Borg I, Groenen P: Modern Multidimensional Scaling: Theory and Applications. Journal of Educational Measurement 2003, 40(3):277-80. 10.1111/j.1745-3984.2003.tb01108.x. Mandal S, Van Treuren W, White RA, Eggesbo M, Knight R, Peddada SD: Analysis of composition of microbiomes: a novel method for studying microbial composition. Microb Ecol Health Dis 2015, 26:27663. doi:10.3402/mehd.v26.27663. Walker AR, Grimes TL, Datta S, Datta S: Unraveling bacterial fingerprints of city subways from microbiome 16S gene profiles. Biology Direct 2018, 13(1):10. 10.1186/s13062-018-0215-8. Walker AR, Datta S: Identification of city specific important bacterial signature for the MetaSUB CAMDA challenge microbiome data. Biology Direct 2019, 14(1):11. 10.1186/s13062-019-0243-z. Kuczynski J, Stombaugh J, Walters WA, Gonzalez A, Caporaso JG, Knight R: Using QIIME to analyze 16S rRNA gene sequences from microbial communities. Curr Protoc Bioinformatics 2011, Chapter 10:Unit 10.7. doi:10.1002/0471250953.bi1007s36. Team RC: R: A language and environment for statistical computing. R foundation for Statistical Computing 2018. Law CW, Chen Y, Shi W, Smyth GK: voom: precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biology 2014, 15(2):R29. doi:10.1186/gb-2014-15-2-r29. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK: limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research 2015, 43(7):e47-e. doi:10.1093/nar/gkv007. Balakrishnama S, Ganapathiraju A: Linear Discriminant Analysis—A Brief Tutorial, vol. 11; 1998. Strobl C, Boulesteix AL, Zeileis A, Hothorn T: Bias in random forest variable importance measures: illustrations, sources and a solution. BMC Bioinformatics 2007, 8:25. doi:10.1186/1471-2105-8-25. Dimitriadou E, Hornik K, Leisch F, Meyer D, Weingessel A: Misc functions of the Department of Statistics (e1071), TU Wien. R package 2008, 1:5-24. Ripley B: MASS: support functions and datasets for Venables and Ripley’s MASS. R package version 2011:7.3-29. Oksanen J, Kindt R, Legendre P, O’Hara B, Stevens MHH, Oksanen MJ, Suggests M: The vegan package. Community ecology package 2007, 10:631-7. Paradis E, Blomberg S, Bolker B, Brown J, Claude J, Cuong HS, Desper R: Package ‘ape’. Analyses of phylogenetics and evolution, version 2019, 2(4). Tables Table 1 . Number of samples included in the analyses and their corresponding city and country of provenance. Table also showed the number of “species”, “family”, and “order” existing in each city. The main dataset­ City Country Number of samples Number of samples used Species Family Order Auckland (AKL) New Zealand 14 14 133 43 21 Berlin (BER) Germany 21 21 235 101 51 Bogota (BOG) Colombia 15 15 703 205 138 Hamilton (HAM) New Zealand 16 16 141 39 22 Hong Kong (HGK) China 18 17 289 112 60 Ilorin (ILR) Nigeria 24 24 230 54 29 London (LON) U.K. 24 22 48 29 15 Marseille (MAR) France 10 10 212 85 44 New York (NYC) U.S.A. 26 26 562 159 83 Offa (OFA) Nigeria 20 20 349 85 42 Porto (PXO) Portugal 20 20 286 122 66 Sacramento (SAC) U.S.A. 18 18 554 208 126 Sao Paulo (SAO) Brazil 24 24 227 99 54 Sofia (SOF) Bulgaria 10 10 275 102 49 Stockholm (STO) Sweden 20 20 186 76 42 Tokyo (TOK) Japan 25 25 573 173 103 All cities - 305 302 1047 276 180 The mystery dataset City Country Number of samples Number of samples used Species Family Order Brisbane (Bri) Australia 7 6 74 53 32 Doha (Doh) Qatar 3 3 59 38 24 Kiev (Kie) Ukraine 8 7 144 80 45 Oslo (Osl) Norway 12 12 505 148 83 Paris (Par) France 8 6 136 73 41 Rio de Janeiro (Rio) Brasil 12 12 343 109 53 Santiago (San) Chile 6 6 309 113 58 Vienna (Vie) Austria 5 5 167 72 35 All cities - 61 57 700 188 109 Table 2. The number of the features selected based on additional rules. The main dataset Rules Number of Features Species Family Order a) Features existing in at least N cities N=15 13 23 17 N=14 26 31 19 N=13 52 43 23 N=12 75 54 29 N=11 110 64 33 N=10 150 73 36 N=9 188 86 43 N=8 234 97 48 b) Top M features with the highest ubiquity across all the samples M=10 10 10 10 M=20 20 20 20 M=30 30 30 30 M=50 50 50 50 M=100 100 100 100 M=150 150 150 150 c) Combination of the common features “species”, “family” and “order” 25 (7 species, 9 families, 9 orders) “species” and “family” 16 (7 species, 9 families) “species” and “order” 16 (7 species, 9 orders) “family” and “order” 18 (9 families, 9 orders) The mystery dataset c) Combination of the common features “species”, “family” and “order” 41 (8 species, 18 families, 15 orders) “species” and “family” 26 (8 species, 18 families) “species” and “order” 23 (8 species, 15 orders) “family” and “order” 33 (18 families, 15 orders) Table 3. The error rate with the leave-one-out cross-validation based on different rules. The number of features selected was retained in brackets. Methods Rules Random Forest Support Vector Machine Linear Discriminant Analysis Species Family Order Species Family Order Species Family Order The main dataset Common features 0.583 (7) 0.299 (9) 0.218 (9) 0.562 (7) 0.276 (9) 0.239 (9) 0.613 (7) 0.303 (9) 0.328 (9) Features existing in at least N cities N=15 0.464 (13) 0.357 (23) 0.369 (17) 0.459 (13) 0.327 (23) 0.344 (17) 0.517 (13) 0.361 (23) 0.381 (17) N=14 0.370 (26) 0.335 (31) 0.349 (19) 0.340 (26) 0.298 (31) 0.339 (19) 0.352 (26) 0.286 (31) 0.384 (19) N=13 0.345 (52) 0.287 (43) 0.313 (23) 0.325 (52) 0.289 (43) 0.291 (23) 0.336 (52) 0.263 (43) 0.289 (23) N=12 0.351 (75) 0.297 (54) 0.299 (29) 0.324 (75) 0.273 (54) 0.292 (29) 0.296 (75) 0.235 (54) 0.243 (29) N=11 0.329 (110) 0.299 (64) 0.291 (33) 0.313 (110) 0.270 (64) 0.281 (33) 0.291 (110) 0.240 (64) 0.216 (33) N=10 0.318 (150) 0.296 (73) 0.290 (36) 0.321 (150) 0.276 (73) 0.266 (36) 0.345 (150) 0.257 (73) 0.218 (36) N=9 0.291 (188) 0.292 (86) 0.299 (43) 0.299 (188) 0.287 (86) 0.283 (43) 0.379 (188) 0.249 (86) 0.210 (43) N=8 0.306 (234) 0.303 (97) 0.280 (48) 0.308 (234) 0.298 (97) 0.268 (48) 0.506 (234) 0.274 (97) 0.203 (48) Top M features with the highest ubiquity across all the samples M=10 0.456 (10) 0.411 (10) 0.425 (10) 0.474 (10) 0.416 (10) 0.433 (10) 0.502 (10) 0.474 (10) 0.447 (10) M=20 0.350 (20) 0.318 (20) 0.333 (20) 0.355 (20) 0.322 (20) 0.328 (20) 0.366 (20) 0.309 (20) 0.325 (20) M=30 0.351 (30) 0.274 (30) 0.289 (30) 0.317 (30) 0.286 (30) 0.275 (30) 0.336 (30) 0.270 (30) 0.239 (30) M=50 0.290 (50) 0.290 (50) 0.269 (50) 0.272 (50) 0.276 (50) 0.250 (50) 0.251 (50) 0.231 (50) 0.198 (50) M=100 0.264 (100) 0.298 (100) 0.282 (100) 0.272 (100) 0.305 (100) 0.301 (100) 0.231 (100) 0.233 (100) 0.272 (100) M=150 0.273 (150) 0.313 (150) 0.296 (150) 0.276 (150) 0.308 (150) 0.398 (150) 0.282 (150) 0.275 (150) 0.398 (150) Combination of the common features 7 species, 9 families, 9 orders 0.116 (25) 0.115 (25) 0.118 (25) 7 species, 9 families 0.248 (16) 0.215 (16) 0.236 (16) 7 species, 9 orders 0.189 (16) 0.189 (16) 0.249 (16) 9 families, 9 orders 0.133 (18) 0.118 (18) 0.133 (18) The mystery dataset Common features 0.598 (8) 0.370 (18) 0.307 (15) 0.615 (8) 0.420 (18) 0.332 (15) 0.659 (8) 0.320 (18) 0.314 (15) Combination of the common features 8 species, 18 families, 15 orders 0.272 (41) 0.327 (41) 0.445 (41) 8 species, 18 families 0.369 (26) 0.47 (26) 0.415 (26) 8 species, 15 orders 0.297 (23) 0.321 (23) 0.282 (23) 18 families, 15 orders 0.255 (33) 0.332 (33) 0.343 (33) Supplementary Files TableS1.xlsx TableS2.xlsx TableS3.xlsx Cite Share Download PDF Status: Published Journal Publication published 04 Jan, 2021 Read the published version in Biology Direct → Version 1 posted Editorial decision: Revision 17 Aug, 2020 Review # 4 received at journal 15 Aug, 2020 Reviewer # 4 agreed at journal 19 Jul, 2020 Review # 2 received at journal 26 Jun, 2020 Review # 3 received at journal 26 Jun, 2020 Review # 1 received at journal 26 Jun, 2020 Reviewer # 3 agreed at journal 10 Jun, 2020 Reviewer # 2 agreed at journal 08 Jun, 2020 Reviewer # 1 agreed at journal 02 Mar, 2020 Reviewers invited by journal 07 Feb, 2020 Editor assigned by journal 21 Jan, 2020 Editor invited by journal 20 Jan, 2020 Submission checks completed at journal 10 Jan, 2020 First submitted to journal 09 Jan, 2020 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-11276","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research","associatedPublications":[],"authors":[{"id":289987,"identity":"35fbb206-eb00-4f87-97bc-ca7c5cb2e7cb","order_by":1,"name":"Runzhi Zhang","email":"","orcid":"https://orcid.org/0000-0002-0763-5928","institution":"University of Florida","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Runzhi","middleName":"","lastName":"Zhang","suffix":""},{"id":289988,"identity":"65484e7d-6fd1-452b-8641-804d9012d395","order_by":2,"name":"Alejandro R. Walker","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Alejandro","middleName":"R.","lastName":"Walker","suffix":""},{"id":289989,"identity":"ea99cf1c-563b-4998-b8a3-fdfcbf6721e3","order_by":3,"name":"Susmita Datta","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7klEQVRIiWNgGAWjYLCCBBBxAERUMPDzASkJwlpAeg4wMzYwnGGQbCNKCwNMC2MbEVrkZ6Q/3fDwhx0D3/Hzxx/8nHdYgo2B+eBtHjxaDG7kmN1ISEhmkDyTzNjYuw2khS3ZGq8WiRw2oBZmBoMDyYwNvNsO17Ex8JhJ49MCdNgzoJZ6BoPzjxkb/84B2cL/Da8WhhsJIIcdBrowmbGZtwGkhYcNrxaDM2+AWtKO80jeeGw4W+ZYugQbM5ux5Rx8DmtPf3bzh021HN/5xAcf39RYS/CzNz+88Qafw6AAySXMRCgfBaNgFIyCUYAfAACYq0spvrEWlQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-7408-699X","institution":"University of Florida","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Susmita","middleName":"","lastName":"Datta","suffix":""}],"badges":[],"createdAt":"2020-01-10 12:32:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.2.20675/v1","doiUrl":"https://doi.org/10.21203/rs.2.20675/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s13062-020-00284-1","type":"published","date":"2021-01-04T15:03:57+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":356044,"identity":"b0311f64-10de-4bd3-be0d-40378b934380","added_by":"auto","created_at":"2020-01-13 17:42:24","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":792841,"visible":true,"origin":"","legend":"Error rate of each city based on combined features for both training datasets: main data dataset in A and mystery dataset in B. The cities are ordered by average error rate of three methods. The features were combined with common “species”, “family”, “order” and common “species”, “order” for main and mystery dataset, respectively.","description":"","filename":"Figure1.jpeg","url":"https://assets-eu.researchsquare.com/files/b846c496-51ab-4247-9e70-af3b5fcd0bee/v1/Figure 1.jpeg"},{"id":356046,"identity":"4c0057c9-d6f1-463b-9782-bbfc10ba20d3","added_by":"auto","created_at":"2020-01-13 17:42:24","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":543634,"visible":true,"origin":"","legend":"The bi-plots of first and second PCoA axes: main dataset in A and mystery dataset in B.","description":"","filename":"Figure2.jpeg","url":"https://assets-eu.researchsquare.com/files/b846c496-51ab-4247-9e70-af3b5fcd0bee/v1/Figure 2.jpeg"},{"id":356048,"identity":"bb12a382-7f61-4faa-b440-0b495cee7a80","added_by":"auto","created_at":"2020-01-13 17:42:24","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1578394,"visible":true,"origin":"","legend":"The analysis of composition of microbiomes across all pair-wise comparisons of cities: main dataset in A and mystery dataset in B. The significant features are denoted by deep blue, the features that are not significantly different in two cities are denoted by light blue.","description":"","filename":"Figure3.jpeg","url":"https://assets-eu.researchsquare.com/files/b846c496-51ab-4247-9e70-af3b5fcd0bee/v1/Figure 3.jpeg"},{"id":356049,"identity":"dff5e4a4-1747-4b2e-878f-7c22b839e4c1","added_by":"auto","created_at":"2020-01-13 17:42:24","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":338243,"visible":true,"origin":"","legend":"The importance of features obtained from the RF. The features are ordered by the importance.","description":"","filename":"Figure4.jpeg","url":"https://assets-eu.researchsquare.com/files/b846c496-51ab-4247-9e70-af3b5fcd0bee/v1/Figure 4.jpeg"},{"id":13485117,"identity":"df67fbc0-69e9-40cd-a469-0a4905935619","added_by":"auto","created_at":"2021-09-16 22:00:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":781595,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-11276/v1/5f8bf13d-5a20-406a-abf2-edf36aff8cf5.pdf"},{"id":356047,"identity":"cbb0bcbd-ba9a-4739-9214-50b860d61c5c","added_by":"auto","created_at":"2020-01-13 17:42:24","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":13022,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/b846c496-51ab-4247-9e70-af3b5fcd0bee/v1/Table S1.xlsx"},{"id":356045,"identity":"982bf78c-097b-4457-83e1-51e5bcfe3e70","added_by":"auto","created_at":"2020-01-13 17:42:24","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":14727,"visible":true,"origin":"","legend":"","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/b846c496-51ab-4247-9e70-af3b5fcd0bee/v1/Table S2.xlsx"},{"id":356043,"identity":"b3f61e21-5d41-419d-97d1-a7af542d18be","added_by":"auto","created_at":"2020-01-13 17:42:23","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":12771,"visible":true,"origin":"","legend":"","description":"","filename":"TableS3.xlsx","url":"https://assets-eu.researchsquare.com/files/b846c496-51ab-4247-9e70-af3b5fcd0bee/v1/Table S3.xlsx"}],"financialInterests":"","formattedTitle":"Unraveling city-specific signature and identifying sample origin locations for the data from CAMDA MetaSUB challenge","fulltext":[{"header":"Background","content":"\u003cp\u003eThe advent of next generation sequencing (NGS) technologies for metagenomics has experienced a tremendous improvement, which allows the generation of large sequence datasets derived from diverse ecosystems, such as the human body, soil, and ocean water [1]. The use of whole genome sequencing (WGS) has been reported to have multiple advantages when compared with the 16S rRNA amplicon data [2]. As the composition of microbial communities can be location specific [3], studying the microbiome from different cities improves our understanding of city-specific microbes and their contributions to ecosystem composition and diversity.\u003c/p\u003e\n\u003cp\u003eIn this work, we aimed to unravel city-specific signature and find the appropriate features for identifying and predicting the origin location of samples from different areas. The dataset was provided by MetaSUB (\u003ca href=\"http://camda2019.bioinf.jku.at/doku.php/contest_dataset\"\u003ehttp://camda2019.bioinf.jku.at/doku.php/contest_dataset\u003c/a\u003e), which aimed to build an international metagenomic map of urban spaces, based on extensive sampling of mass-transit system and other public areas around the world. They partnered with CAMDA for an early release of microbiome data obtained from global City Sampling Days, comprising the WGS metagenomics data. The main dataset covered 16 cities across the globe, with tens of samples per city. Moreover, one more dataset with 8 cities was provided as mystery set from the CAMDA 2019 MetaSUB challenge to serve as testing samples. And the true city-information of the mystery data was provided much later in the process. \u003cstrong\u003eTable 1\u003c/strong\u003e presented a tabulated insight of the data for all the cities.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eFeature selection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFirst, we selected the common \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo;, and \u0026ldquo;order\u0026rdquo; existing across all the 16 cities and the number of variables was 7, 9 and 9 respectively. The number of features was limited and more information about the microbes was needed. In order to increase the number of features, we selected the features based on additional rules: a) The features were selected based on the ubiquity of the \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; across all the cities. b) The features were selected based on the ubiquity of the \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; across all the samples. c) The combinations of the common \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo;, and \u0026ldquo;order\u0026rdquo; were regarded as the combined features. \u003cstrong\u003eTable 2\u003c/strong\u003e presented the details of the features selected based on additional rules. For simplicity, the mystery dataset was analyzed based on the common features and combined features. After feature selection, the aggregated raw counts of each dataset were normalized to generate log2-cpm for further analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMachine learning analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the main and mystery datasets respectively, results from Random forest (RF) [4], Support Vector Machine (SVM) [5] and Linear Discriminant Analysis (LDA) [6] were obtained with the leave-one-out cross validation (CV), one test sample was randomly selected in each run with 1000 runs repeated. \u003cstrong\u003eTable 3\u003c/strong\u003e presented the details of the classification error rate based on different rules using the main and mystery datasets respectively.\u003c/p\u003e\n\u003cp\u003eAs seen in the table, when the features existing in at least N cities were selected for the analysis, the changing trends of the error rate of RF-species (qualitied \u0026ldquo;species\u0026rdquo; was used for analysis using RF), SVM-species, LDA-species, LDA-family and LDA-order shared a similar pattern, a decreased CV error rate was obtained when increasing the number of features, and then the lowest error rate was achieved. For RF-family and SVM-family, the error rate hasn\u0026rsquo;t changed considerably when we increased the number of the variables at the family rank. Therefore, using the common \u0026ldquo;family\u0026rdquo; was better, as we obtained the low error rate without including too many features. And for RF-order and SVM-order, the lowest error rate was obtained using common \u0026ldquo;order\u0026rdquo;. Additionally, when top M features with the highest ubiquity across all the samples were selected for analysis, error rates decreased with the increasing number of features used and then the lowest error rate was achieved, no matter which machine learning methods or which kinds of feature we used for analysis. In addition, for the combined features, the best performance was achieved using the features combined with common \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; (7 species, 9 families, 9 orders), the error rates obtained from RF, SVM and LDA were 11.6%, 11.5% and 11.8% respectively. And for the mystery dataset, we obtained the lowest average error rate when we used the combined features with 8 species and 15 orders. The error rates were 29.7%, 32.1% and 28.2% for RF, SVM and LDA respectively.\u003c/p\u003e\n\u003cp\u003eThe error rates of predictions for each city was presented in \u003cstrong\u003eFigure 1\u003c/strong\u003e. From \u003cstrong\u003eFigure 1.A\u003c/strong\u003e, the low error rates were obtained from the cities with better sequencing depth (\u003cstrong\u003eTable S1\u003c/strong\u003e) including Bogota, Ilorin, New York, Offa, Sacramento and Tokyo. The average error rates of the three methods for these four cities were 7.84%, 5.49%, 2.03%, 6.31%, 3.08% and 5.80% respectively. However, the error rates of Auckland and Hamilton, two cities also with good sequencing depth, were 26.39% and 18.44% respectively. By looking into the details of the results, we found that the samples AKL_1, AKL_7, AKL_14 and HAM_7, HAM_12 cannot be predicted correctly by all the three methods. In other words, the microbial composition of these samples was different from the other samples in Auckland and Hamilton (\u003cstrong\u003eTable S2\u003c/strong\u003e), making them difficult to be identified. And we found that the samples from London with the poor sequencing depth showed low error rate. Upon finding excessive zeros in samples from London, the samples of London could be easily identified from all the samples, which resulted in the low error rate of London. In addition, a possible explanation for low error rate could be the insufficient number of samples, as the cities with the lowest number of samples including Sofia and Marseille showed high error rates. And from \u003cstrong\u003eFigure 1.B\u003c/strong\u003e, the cities with deep sequencing (\u003cstrong\u003eTable S3\u003c/strong\u003e), i.e. Oslo and Rio de Janeiro, were the two of the first three cities with the lowest average error rate (6.64% and 16.81% respectively). Similarly, the cities with poor sequencing depth such as Doha and Brisbane, showed the high average error rate (85.71% and 62.61% respectively). And Doha, also with the limited number of samples, achieved the highest error rate among all the cities. The evident from the \u003cstrong\u003eFigure 1\u003c/strong\u003e that most cities with limited samples and poor sequencing depth had high error rates, indicating that sufficient samples and deep sequencing were necessary for successfully predicting the provenance of samples.\u003c/p\u003e\n\u003cp\u003eIn addition to the analyses based on the main and mystery datasets respectively, we have also used the prediction models built based on the main dataset to predict the samples from mystery dataset. As the main dataset and mystery dataset had no cities in common, the information of cities from mystery dataset was lacked in the main dataset. Therefore, 50% of the samples of each city from mystery dataset were randomly sampled and added to the main dataset to serve as the part of the training dataset, and the rest of the samples from mystery dataset were served as the test samples. In the training dataset, the samples from mystery dataset were labeled as the \u0026ldquo;mystery\u0026rdquo;. The prediction models were built based on the common \u0026ldquo;family\u0026rdquo; (5 families) and \u0026ldquo;order\u0026rdquo; (6 orders), as there was no common \u0026ldquo;species\u0026rdquo; between the main and mystery datasets. Three different classifications were used and the random samplings were conducted for 1000 times. For each run, the predicted labels were checked with the real labels. The average error rates for RF, SVM and LDA were 10.48%, 9.21% and 10.36% respectively, indicating that the prediction models could effectively identify the mystery samples from all the samples.\u003c/p\u003e\n\u003cp\u003eThe following analyses were based on the features which achieved the lowest average error rate.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePrincipal Coordinates Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe results of PCoA [7] in \u003cstrong\u003eFigure 2\u003c/strong\u003e presented the bi-plots for both datasets. \u003cstrong\u003eFigure 2.A\u003c/strong\u003e illustrated the main dataset and the 58.4% of total variability of the data could be explained by the first two PCoA axes. A separation of the cities could be referred from the plot. For example, London was separated from most cities and on the rightmost site, which was corresponding to the result from machine learning methods, indicating that the excessive zeros in samples from London made the prediction easier. However, many cities overlapped together. Specifically, Ilorin and Offa, both the cities of Nigeria, whose ellipses showed a massive overlap. Also, Auckland and Hamilton, both being in New Zealand, overlapped with each other. The result of mystery dataset was given in \u003cstrong\u003eFigure 2.B\u003c/strong\u003e. The first two PCoA axes explained 65.4% of total variability in the data, which was comparable with the percentage explained in the main dataset. Although many cities overlapped, samples of Oslo were clustered together and distributed at the top of the plot, separating from the most samples. Also, some of the samples from Rio de Janeiro were far away from the most cities, which made these samples easier to be identified. These results were corresponding to the low error rates of Oslo and Rio de Janeiro in the previous section.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnalysis of composition of microbiomes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe results from the analysis of composition of microbiomes (ANCOM) [8] were presented in \u003cstrong\u003eFigure 3\u003c/strong\u003e. The relative abundances of the features were used to conduct the pair-wise comparisons among all the cities. Upon the significance of the features, the differentially abundant features were found. The features, which served as the predictors, on the right were ordered by the number of times the relative abundance was significantly different in the pair-wise comparisons. As presented in \u003cstrong\u003eFigure 3.A\u003c/strong\u003e, the top 10 features, i.e. \u003cem\u003eBacillaceae\u003c/em\u003e,\u003cem\u003e Bacillales\u003c/em\u003e, \u003cem\u003eActinomycetales\u003c/em\u003e, \u003cem\u003eSphingomonadaceae\u003c/em\u003e, \u003cem\u003ePseudomonadaceae\u003c/em\u003e, \u003cem\u003ePseudomonas.spp\u003c/em\u003e, \u003cem\u003eSphingomonadales\u003c/em\u003e, \u003cem\u003eLactobacillales\u003c/em\u003e, \u003cem\u003estreptococcaceae\u003c/em\u003e and \u003cem\u003eEnterobacteriaceae\u003c/em\u003e showed the highest counts. For the top feature, the count of \u003cem\u003eBacillaceae\u003c/em\u003e is 43 out of 120, which meant that in the 120 pair-wise comparisons among the 16 cities, \u003cem\u003eBacillaceae\u003c/em\u003e was found to be significantly different in 43 comparisons. Similarly, in \u003cstrong\u003eFigure 3.B\u003c/strong\u003e, the top 10 features were \u003cem\u003eBacillales\u003c/em\u003e, \u003cem\u003eClostridiales\u003c/em\u003e, \u003cem\u003ePseudomonadales\u003c/em\u003e, \u003cem\u003eStaphylococcus.epidermidis\u003c/em\u003e, \u003cem\u003eLactobacillales\u003c/em\u003e, \u003cem\u003eRhodospirillales\u003c/em\u003e, \u003cem\u003eFlavobacteriales\u003c/em\u003e, \u003cem\u003eStreptophyta\u003c/em\u003e, \u003cem\u003eBurkholderiales\u003c/em\u003e and \u003cem\u003eEnterobacteriales\u003c/em\u003e. Additionally, little change was seen from the importance score (\u003cstrong\u003eFigure 4\u003c/strong\u003e) derived from the RF. It could be inferred from \u003cstrong\u003eFigure 4.A\u003c/strong\u003e that \u003cem\u003eBacillales, Actinomycetales\u003c/em\u003e, \u003cem\u003eSphingomonadales\u003c/em\u003e, \u003cem\u003eSphingomonadaceae\u003c/em\u003e,\u003cem\u003e Enterobacteriaceae \u003c/em\u003eand\u003cem\u003e streptococcaceae \u003c/em\u003ewere also in top 10 features. Also, for the mystery dataset, \u003cem\u003eBacillales\u003c/em\u003e, \u003cem\u003eStreptophyta, Rhodospirillales\u003c/em\u003e, \u003cem\u003eEnterobacteriales\u003c/em\u003e, \u003cem\u003eLactobacillales\u003c/em\u003e,\u003cem\u003e Clostridiales \u003c/em\u003eand \u003cem\u003ePseudomonadales\u003c/em\u003e were in the top 10 of both lists. In summary, for the main dataset, the common \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; of the OTU provided the most informative data for predicting the origins of the samples, which was also corresponding to the low error rate of three methods when using the common \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; only compared with using the common \u0026ldquo;species\u0026rdquo;. For the mystery dataset, the results of the ANCOM and feature importance combined with the error rates from three methods indicated that the common \u0026ldquo;order\u0026rdquo; dominated the prediction.\u003c/p\u003e"},{"header":"Discussion and conclusions","content":"\u003cp\u003eFor the CAMDA challenge MetaSUB data of this year, 16 cities were included and 10 or more samples were collected for each city in the main dataset. The feature selection, normalization, three methods of machine learning algorithms, PCoA and ANCOM were conducted for both the main and mystery datasets. Combining the feature selection and the results of machine learning methods, we found that the machine learning analysis was effective in predicting the origin of the samples when the combined common features were used, indicating that the combination of the common features could be the effective microbial fingerprint for unraveling city-specific signature and identifying sample origin locations, which proved to be with great potential for forensic science. Therefore, more taxonomic ranks of microbiomes such as class and genus could be added to the combination to investigate the performance of the prediction in the future works. As in this study, by combining the common features, we obtained the low error rate of prediction without including too many features for both datasets. Additionally, we have used the main dataset and half of mystery samples as our training dataset to predict the labels of another half of mystery samples. As the error rates were 10.48%, 9.21% and 10.36% respectively for RF, SVM and LDA, most of the samples from mystery dataset could be identified correctly by the classification. PCoA analysis for both datasets showed that nearly 60% of the total variability of the data could be explained by the first two PCoA axes, most cities overlapped with each other. However, some cities such as London and Oslo were separated from most cities, indicating the unique composition of microbiomes in these cities, which was also validated by ANCOM analysis and was corresponding to the low error rates in machine learning analysis. The heatmaps of ANCOM revealed that some of the features, such as some of the common \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; in the main dataset and the common \u0026ldquo;order\u0026rdquo; in mystery dataset, were significantly different in pair-wise comparisons, and these features were also given high importance score in RF, indicating the effectiveness of these features during the classification. Additionally, this was also supported by the results of the machine learning analysis, we obtained a lower error rate using common \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; and using common \u0026ldquo;order\u0026rdquo; only for the main and mystery datasets respectively.\u003c/p\u003e\n\u003cp\u003eHowever, due to the limited resources such as the insufficient number of samples from some cities and the poor sequencing depth of the samples, the error rate of Sofia was higher compared with other cities. Therefore, sufficient samples and better sequencing depth were necessary in order to improve the accuracy of the prediction. A similar problem was found in the mystery dataset, the number of samples was limited for most of the cities in the mystery dataset, which resulted in an overall higher error rate in the mystery dataset compared with the main dataset. And the poor sequencing of Brisbane and Doha also resulted in higher error rates compared to other cities in the mystery dataset. Additionally, the microbial composition could vary across samples within the same city, which also limited the ability of classification to correctly predict the origins of the samples. For this study, only the combined common features were selected for further analysis, some city-specific microbiomes might be ignored in this study, as they only existed in a specific city. These city-specific microbiomes could be combined with the common features to provide more powerful prediction in the future work.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eThe design of the analysis was motivated by the experience from the CAMDA 2017 and CAMDA 2018 MetaSUB Challenges [9, 10]. Compared to date from previous MeteSUB challenges, the data this year was with higher quality and deeper sequencing depth. As we have more cities included this year, the common features shared by all the cities were further limited. Therefore, the features selection was implemented this year to help us obtain the appropriate features for prediction. Finally, unsupervised and supervised techniques were used for the analyses. A more detailed description of the implementations was supplemented in the following sections.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBioinformatics and data preparation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Bioinformatics and data preparation was based on our previous paper [10]. Samples were Illumina-sequenced at different depths and delivered as FASTQ format for further analysis. Subsequent bioinformatics processing and data preparation were conducted in the \u0026ldquo;HyperGator2\u0026rdquo; high-performance cluster at the University of Florida. After quality control, the data was picked for OTUs in open-reference mode with QIIME [11]. After quality control, samples with the poor sequencing depth were removed. OTUs were aggregated as counts, and all further data processing and analysis in this study were conducted in R [12].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFeature selection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFeature selection was conducted based on different rules. For the main dataset, the common \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; existing across all the 16 cities were selected at first. And three additional rules were conducted for feature selection: a) \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; existing in at least N cities were selected. N was set to 15, 14, 13, 12, 11, 10, 9 and 8 respectively, the count of the feature that did not exist in the city was marked as zero. b) \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo; were reordered based on their ubiquity across all the samples respectively, the top M features with the highest ubiquity were selected. M was set to 10, 20, 30, 50, 100, 150 respectively. c) The combinations of the common \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo;, and \u0026ldquo;order\u0026rdquo; were regarded as the combined features. The mystery dataset was analyzed based on the common features and combined features. The data were then normalized to generate log2-cpm to make the data meaningful using the function \u0026ldquo;voom\" [13] in R package \u0026ldquo;limma\u0026rdquo; [14] for further analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMachine learning analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThree classification algorithms were implemented at this stage for both the main and mystery datasets: Random Forest (RF) [4], Support Vector Machine (SVM) [5] and Linear Discriminant Analysis (LDA) [15]. When the analysis was conducted for the main and mystery datasets respectively, each method was implemented 1000 times with the leave-one-out cross validation based on the selected features. For each run, one sample was randomly selected as the test data with other samples served as the training set. RF was conducted using the R package \u0026ldquo;randomForest\u0026rdquo;, 1000 trees were used and the count of variables chosen at each split was equivalent to the square root of the number of features in the dataset. The model was fitted based on the training set and then used to predict the origin of the test sample. The result of each prediction was obtained. The overall error rate and the error rate for each city were calculated based on the result of the 1000 runs. The variable importance score [16] computed by the RF was also recorded. The SVM classifier was implemented in a similar manner. The two important parameters in SVM, i.e. gamma and c-value, affected the fitting of the models. Using the R function \u0026ldquo;best.svm\u0026rdquo; in package \u0026ldquo;e1071\u0026rdquo; [17], the SVM model with the appropriate parameters was obtained by testing the performance of models with different parameters. The LDA was conducted in a similar manner using the R package \u0026ldquo;MASS\u0026rdquo; [18].\u003c/p\u003e\n\u003cp\u003eWhen the models based on the main dataset were used to predict the labels of samples from mystery dataset. 50% of the samples of each city were randomly sampled from mystery dataset and added to the main dataset with the labels of \u0026ldquo;mystery\u0026rdquo; to serve as the part of the training dataset, the rest of the samples were served as the test samples. The sampling and the prediction were implemented for 1000 times. For each run, the predicted labels of test samples were checked with the real labels, a good prediction meant all test samples could be labeled as \u0026ldquo;mystery\u0026rdquo; by the classification.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePrincipal Coordinates analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrincipal coordinates analysis (PCoA) [7] of normalized data was conducted using the R package \u0026ldquo;vegan\u0026rdquo; [19] and \u0026ldquo;ape\u0026rdquo; [20]. Firstly, the dissimilarity matrix was constructed based on Bray-Cruits distance. Then, the dissimilarity matrix was used for PCoA and a set of uncorrelated axes was generated to summarise the variability in the dataset. The two-dimensional plot was generated for assessing the separation of the cities.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnalysis of composition of microbiomes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnalysis of composition of microbiomes for the normalized data was conducted using the R package \u0026ldquo;ANCOM\u0026rdquo; [8], which accounted for the underlying structure in the microbial data and was used to comparing the composition of microbiomes in two or more populations. For the main and mystery datasets, ANCOM across all pair-wise comparisons in each dataset was implemented. The 120 and 28 pair-wise comparisons were made for the combination of all cities in the main and mystery datasets respectively. And the output of the ANCOM was a set of differentially abundant features between the two cities for each comparison at the significance level of 0.05. The heatmap was made based on the output for investigating the difference of microbial composition between different cities. Additionally, the results of ANCOM were compared with the importance score derived by the RF method.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets supporting the conclusions of this article can be obtained from the CAMDA 2019 website \u003ca href=\"http://camda2019.bioinf.jku.at/doku.php/contest_dataset\"\u003ehttp://camda2019.bioinf.jku.at/doku.php/contest_dataset\u003c/a\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDatta, S. was partially supported by NIH grant 1UL1TR000064 from the National Center for Advancing Translational Sciences.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor\u0026rsquo;s contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSD reviewed the manuscript and provided theoretical support when required, RZ and ARW designed, run the analyses, RZ wrote the manuscript. All the authors have read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe samples were provided to the CAMDA 2019 competition by the MetaSUB Consortium.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e1\u003c/sup\u003eDepartment of Biostatistics, University of Florida, 2004 Mowry Rd, Gainesville, FL, 32610, USA\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e2\u003c/sup\u003eDepartment of Oral Biology, University of Florida, 1395 Center Drive, Gainesville, FL, 32610, USA\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eNGS: Next Generation Sequencing\u003c/p\u003e\n\u003cp\u003eWGS: whole genome sequencing\u003c/p\u003e\n\u003cp\u003eOTU: Operational Taxonomic Unit\u003c/p\u003e\n\u003cp\u003eRF: Random Forest\u003c/p\u003e\n\u003cp\u003eSVM: Support Vector Machine\u003c/p\u003e\n\u003cp\u003eLDA: Linear Discriminant Analysis\u003c/p\u003e\n\u003cp\u003eCV: Cross Validation\u003c/p\u003e\n\u003cp\u003ePCoA: Principal Coordinates Analysis\u003c/p\u003e\n\u003cp\u003eANCOM: Analysis of composition of microbiomes\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSimon C, Daniel R: Metagenomic analyses: past and future trends. Appl Environ Microbiol 2011, 77(4):1153-61. doi:10.1128/aem.02345-10.\u003c/li\u003e\n\u003cli\u003eRanjan R, Rani A, Metwally A, McGee HS, Perkins DL: Analysis of the microbiome: Advantages of whole genome shotgun versus 16S amplicon sequencing. Biochem Biophys Res Commun 2016, 469(4):967-77. doi:10.1016/j.bbrc.2015.12.083.\u003c/li\u003e\n\u003cli\u003eDelgado-Baquerizo M, Oliverio AM, Brewer TE, Benavent-Gonz\u0026aacute;lez A, Eldridge DJ, Bardgett RD, Maestre FT, Singh BK, Fierer N: A global atlas of the dominant bacteria found in soil. Science 2018, 359(6373):320-5. doi:10.1126/science.aap9516.\u003c/li\u003e\n\u003cli\u003eBreiman L: Random Forests. Machine Learning 2001, 45(1):5-32. doi:10.1023/a:1010933404324.\u003c/li\u003e\n\u003cli\u003eCortes C, Vapnik V: Support-vector networks. Machine Learning 1995, 20(3):273-97. doi:10.1007/bf00994018.\u003c/li\u003e\n\u003cli\u003eBalakrishnama S, Ganapathiraju A: Linear discriminant analysis-a brief tutorial. Institute for Signal and information Processing 1998, 18:1-8.\u003c/li\u003e\n\u003cli\u003eBorg I, Groenen P: Modern Multidimensional Scaling: Theory and Applications. Journal of Educational Measurement 2003, 40(3):277-80. 10.1111/j.1745-3984.2003.tb01108.x.\u003c/li\u003e\n\u003cli\u003eMandal S, Van Treuren W, White RA, Eggesbo M, Knight R, Peddada SD: Analysis of composition of microbiomes: a novel method for studying microbial composition. Microb Ecol Health Dis 2015, 26:27663. doi:10.3402/mehd.v26.27663.\u003c/li\u003e\n\u003cli\u003eWalker AR, Grimes TL, Datta S, Datta S: Unraveling bacterial fingerprints of city subways from microbiome 16S gene profiles. Biology Direct 2018, 13(1):10. 10.1186/s13062-018-0215-8.\u003c/li\u003e\n\u003cli\u003eWalker AR, Datta S: Identification of city specific important bacterial signature for the MetaSUB CAMDA challenge microbiome data. Biology Direct 2019, 14(1):11. 10.1186/s13062-019-0243-z.\u003c/li\u003e\n\u003cli\u003eKuczynski J, Stombaugh J, Walters WA, Gonzalez A, Caporaso JG, Knight R: Using QIIME to analyze 16S rRNA gene sequences from microbial communities. Curr Protoc Bioinformatics 2011, Chapter 10:Unit 10.7. doi:10.1002/0471250953.bi1007s36.\u003c/li\u003e\n\u003cli\u003eTeam RC: R: A language and environment for statistical computing. R foundation for Statistical Computing 2018.\u003c/li\u003e\n\u003cli\u003eLaw CW, Chen Y, Shi W, Smyth GK: voom: precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biology 2014, 15(2):R29. doi:10.1186/gb-2014-15-2-r29.\u003c/li\u003e\n\u003cli\u003eRitchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK: limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research 2015, 43(7):e47-e. doi:10.1093/nar/gkv007.\u003c/li\u003e\n\u003cli\u003eBalakrishnama S, Ganapathiraju A: Linear Discriminant Analysis\u0026mdash;A Brief Tutorial, vol. 11; 1998.\u003c/li\u003e\n\u003cli\u003eStrobl C, Boulesteix AL, Zeileis A, Hothorn T: Bias in random forest variable importance measures: illustrations, sources and a solution. BMC Bioinformatics 2007, 8:25. doi:10.1186/1471-2105-8-25.\u003c/li\u003e\n\u003cli\u003eDimitriadou E, Hornik K, Leisch F, Meyer D, Weingessel A: Misc functions of the Department of Statistics (e1071), TU Wien. R package 2008, 1:5-24.\u003c/li\u003e\n\u003cli\u003eRipley B: MASS: support functions and datasets for Venables and Ripley\u0026rsquo;s MASS. R package version 2011:7.3-29.\u003c/li\u003e\n\u003cli\u003eOksanen J, Kindt R, Legendre P, O\u0026rsquo;Hara B, Stevens MHH, Oksanen MJ, Suggests M: The vegan package. Community ecology package 2007, 10:631-7.\u003c/li\u003e\n\u003cli\u003eParadis E, Blomberg S, Bolker B, Brown J, Claude J, Cuong HS, Desper R: Package \u0026lsquo;ape\u0026rsquo;. Analyses of phylogenetics and evolution, version 2019, 2(4).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eTable 1\u003c/span\u003e\u003c/strong\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e. Number of samples included in the analyses and their corresponding city and country of provenance. Table also showed the number of \u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo;, and \u0026ldquo;order\u0026rdquo; existing in each city.\u003c/span\u003e\u003c/p\u003e\n\u003ctable style=\"border-collapse: collapse; border: none;\" width=\"568\"\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 425.65pt; border: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"7\" width=\"568\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 12.0pt; font-family: 'Cambria',serif;\"\u003eThe main dataset\u0026shy;\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eCity\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eCountry\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNumber of samples\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNumber of samples used\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eSpecies\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eFamily\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eOrder\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eAuckland (AKL)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNew Zealand\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e14\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e14\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e133\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e43\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e21\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eBerlin (BER)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eGermany\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e21\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e21\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e235\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e101\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e51\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eBogota (BOG)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eColombia\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e15\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e15\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e703\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e205\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e138\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eHamilton (HAM)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNew Zealand\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e16\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e16\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e141\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e39\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e22\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eHong Kong (HGK)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eChina\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e18\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e17\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e289\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e112\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e60\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eIlorin (ILR)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNigeria\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e24\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e24\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e230\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e54\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e29\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eLondon (LON)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eU.K.\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e24\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e22\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e48\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e29\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e15\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eMarseille (MAR)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eFrance\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e212\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e85\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e44\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNew York (NYC)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eU.S.A.\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e26\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e26\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e562\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e159\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e83\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eOffa (OFA)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNigeria\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e349\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e85\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e42\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003ePorto (PXO)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003ePortugal\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e286\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e122\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e66\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eSacramento (SAC)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eU.S.A.\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e18\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e18\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e554\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e208\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e126\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eSao Paulo (SAO)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eBrazil\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e24\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e24\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e227\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e99\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e54\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eSofia (SOF)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eBulgaria\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e275\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e102\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e49\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eStockholm (STO)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eSweden\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e186\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e76\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e42\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eTokyo (TOK)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eJapan\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e25\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e25\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e573\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e173\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e103\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eAll cities\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e-\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e305\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e302\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e1047\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e276\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e180\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 425.65pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"7\" width=\"568\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 12.0pt; font-family: 'Cambria',serif;\"\u003eThe mystery dataset\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eCity\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eCountry\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNumber of samples\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNumber of samples used\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eSpecies\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eFamily\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eOrder\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eBrisbane (Bri)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eAustralia\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e7\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e6\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e74\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e53\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e32\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eDoha (Doh)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eQatar\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e3\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e3\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e59\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e38\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e24\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eKiev (Kie)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eUkraine\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e8\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e7\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e144\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e80\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e45\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eOslo (Osl)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eNorway\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e12\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e12\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e505\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e148\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e83\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eParis (Par)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eFrance\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e8\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e6\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e136\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e73\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e41\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eRio de Janeiro (Rio)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eBrasil\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e12\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e12\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e343\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e109\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e53\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eSantiago (San)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eChile\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e6\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e6\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e309\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e113\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e58\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eVienna (Vie)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eAustria\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e5\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e5\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e167\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e72\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e35\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 70.75pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"94\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eAll cities\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 55.45pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"74\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e-\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 79.1pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"105\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e61\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 99.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"132\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e57\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e700\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 37.05pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"49\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e188\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 41.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"55\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003e109\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 12.0pt; font-family: 'Cambria',serif;\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eTable 2.\u003c/span\u003e\u003c/strong\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e The number of the features selected based on additional rules. \u003c/span\u003e\u003c/p\u003e\n\u003ctable style=\"border-collapse: collapse; border: none;\"\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 414.8pt; border: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"5\" width=\"553\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 12.0pt; font-family: 'Cambria',serif;\"\u003eThe main dataset\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 226.55pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"2\" rowspan=\"2\" width=\"302\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eRules\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 188.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"251\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eNumber of Features\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eSpecies\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eFamily\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eOrder\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 110.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" rowspan=\"8\" width=\"148\"\u003e\n\u003cp style=\"margin-left: .25in; text-indent: -.25in;\"\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003ea)\u003cspan style=\"font: 7.0pt 'Times New Roman';\"\u003e\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp; \u003c/span\u003e\u003c/span\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eFeatures existing in at least N cities\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eN=15\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e13\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e23\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e17\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eN=14\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e26\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e31\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e19\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eN=13\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e52\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e43\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e23\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eN=12\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e75\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e54\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e29\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eN=11\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e110\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e64\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e33\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eN=10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e150\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e73\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e36\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eN=9\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e188\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e86\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e43\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eN=8\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e234\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e97\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e48\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 110.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" rowspan=\"6\" width=\"148\"\u003e\n\u003cp style=\"margin-left: .25in; text-indent: -.25in;\"\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eb)\u003cspan style=\"font: 7.0pt 'Times New Roman';\"\u003e\u0026nbsp;\u0026nbsp;\u0026nbsp; \u003c/span\u003e\u003c/span\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eTop M features with the highest ubiquity across all the samples\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eM=10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eM=20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eM=30\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e30\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e30\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e30\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eM=50\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e50\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e50\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e50\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eM=100\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e100\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e100\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e100\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eM=150\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e150\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 63.8pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"85\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e150\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 60.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"81\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e150\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 110.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" rowspan=\"4\" width=\"148\"\u003e\n\u003cp style=\"margin-left: .25in; text-indent: -.25in;\"\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003ec)\u003cspan style=\"font: 7.0pt 'Times New Roman';\"\u003e\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp; \u003c/span\u003e\u003c/span\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eCombination of the common features\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e\u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo;\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 188.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"251\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e25 (7 species, 9 families, 9 orders)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e\u0026ldquo;species\u0026rdquo; and \u0026ldquo;family\u0026rdquo;\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 188.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"251\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e16 (7 species, 9 families)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e\u0026ldquo;species\u0026rdquo; and \u0026ldquo;order\u0026rdquo;\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 188.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"251\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e16 (7 species, 9 orders)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e\u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo;\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 188.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"251\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e18 (9 families, 9 orders)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 414.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"5\" width=\"553\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 12.0pt; font-family: 'Cambria',serif;\"\u003eThe mystery dataset\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 110.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" rowspan=\"4\" width=\"148\"\u003e\n\u003cp style=\"margin-left: .25in; text-indent: -.25in;\"\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003ec)\u003cspan style=\"font: 7.0pt 'Times New Roman';\"\u003e\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp; \u003c/span\u003e\u003c/span\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eCombination of the common features\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e\u0026ldquo;species\u0026rdquo;, \u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo;\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 188.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"251\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e41 (8 species, 18 families, 15 orders)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e\u0026ldquo;species\u0026rdquo; and \u0026ldquo;family\u0026rdquo;\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 188.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"251\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e26 (8 species, 18 families)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e\u0026ldquo;species\u0026rdquo; and \u0026ldquo;order\u0026rdquo;\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 188.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"251\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e23 (8 species, 15 orders)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 115.75pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"154\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e\u0026ldquo;family\u0026rdquo; and \u0026ldquo;order\u0026rdquo;\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 188.25pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"251\"\u003e\n\u003cp\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e33 (18 families, 15 orders)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 12.0pt; font-family: 'Cambria',serif;\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003eTable 3.\u003c/span\u003e\u003c/strong\u003e\u003cspan style=\"font-family: 'Cambria',serif;\"\u003e The error rate with the leave-one-out cross-validation based on different rules. The number of features selected was retained in brackets.\u003c/span\u003e\u003c/p\u003e\n\u003ctable style=\"border-collapse: collapse; border: none;\" width=\"553\"\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" rowspan=\"2\" width=\"79\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp; \u003c/span\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 5.5pt; font-family: 'Cambria',serif;\"\u003e\u0026nbsp;Methods\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 5.5pt; font-family: 'Cambria',serif;\"\u003eRules\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 121.65pt; border: solid windowtext 1.0pt; border-left: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"162\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eRandom Forest\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border: solid windowtext 1.0pt; border-left: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eSupport Vector Machine\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border: solid windowtext 1.0pt; border-left: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 7.5pt; font-family: 'Cambria',serif;\"\u003eLinear Discriminant Analysis\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 8.0pt; font-family: 'Cambria',serif;\"\u003eSpecies\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 8.0pt; font-family: 'Cambria',serif;\"\u003eFamily\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 8.0pt; font-family: 'Cambria',serif;\"\u003eOrder\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 8.0pt; font-family: 'Cambria',serif;\"\u003eSpecies\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 8.0pt; font-family: 'Cambria',serif;\"\u003eFamily\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 8.0pt; font-family: 'Cambria',serif;\"\u003eOrder\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 8.0pt; font-family: 'Cambria',serif;\"\u003eSpecies\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 8.0pt; font-family: 'Cambria',serif;\"\u003eFamily\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 8.0pt; font-family: 'Cambria',serif;\"\u003eOrder\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 414.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"10\" width=\"553\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 12.0pt; font-family: 'Cambria',serif;\"\u003eThe main dataset\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 6.5pt; font-family: 'Cambria',serif;\"\u003eCommon features\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.583 (7)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.299 (9)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.218 (9)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.562 (7)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.276 (9)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.239 (9)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.613 (7)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.303 (9)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.328 (9)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 414.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"10\" width=\"553\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 6.5pt; font-family: 'Cambria',serif;\"\u003eFeatures existing in at least N cities\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eN=15\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.464 (13)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.357 (23)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.369 (17)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.459 (13)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.327 (23)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.344 (17)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.517 (13)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.361 (23)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.381 (17)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eN=14\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.370 (26)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.335 (31)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.349 (19)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.340 (26)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.298 (31)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.339 (19)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.352 (26)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.286 (31)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.384 (19)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eN=13\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.345 (52)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.287 (43)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.313 (23)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.325 (52)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.289 (43)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.291 (23)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.336 (52)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.263 (43)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.289 (23)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eN=12\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.351 (75)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.297 (54)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.299 (29)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.324 (75)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.273 (54)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.292 (29)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.296 (75)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.235 (54)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.243 (29)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eN=11\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.329 (110)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.299 (64)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.291 (33)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.313 (110)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.270 (64)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.281 (33)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.291 (110)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.240 (64)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.216 (33)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eN=10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.318 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.296 (73)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.290 (36)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.321 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.276 (73)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.266 (36)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.345 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.257 (73)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.218 (36)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eN=9\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.291 (188)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.292 (86)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.299 (43)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.299 (188)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.287 (86)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.283 (43)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.379 (188)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.249 (86)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.210 (43)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eN=8\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.306 (234)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.303 (97)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.280 (48)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.308 (234)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.298 (97)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.268 (48)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.506 (234)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.274 (97)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.203 (48)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 414.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"10\" width=\"553\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 6.5pt; font-family: 'Cambria',serif;\"\u003eTop M features with the highest ubiquity across all the samples\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eM=10\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.456 (10)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.411 (10)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.425 (10)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.474 (10)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.416 (10)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.433 (10)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.502 (10)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.474 (10)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.447 (10)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eM=20\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.350 (20)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.318 (20)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.333 (20)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.355 (20)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.322 (20)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.328 (20)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.366 (20)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.309 (20)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.325 (20)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eM=30\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.351 (30)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.274 (30)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.289 (30)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.317 (30)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.286 (30)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.275 (30)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.336 (30)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.270 (30)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.239 (30)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eM=50\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.290 (50)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.290 (50)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.269 (50)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.272 (50)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.276 (50)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.250 (50)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.251 (50)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.231 (50)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.198 (50)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eM=100\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.264 (100)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.298 (100)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.282 (100)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.272 (100)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.305 (100)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.301 (100)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.231 (100)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.233 (100)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.272 (100)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: right;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003eM=150\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.273 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.313 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.296 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.276 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.308 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.398 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.282 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.275 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.398 (150)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 414.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"10\" width=\"553\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 6.5pt; font-family: 'Cambria',serif;\"\u003eCombination of the common features \u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e7 species, 9 families, 9 orders\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 121.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"162\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.116 (25) \u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.115 (25)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.118 (25)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e7 species, 9 families\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 121.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"162\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.248 (16)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.215 (16)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.236 (16)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e7 species, 9 orders\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 121.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"162\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.189 (16)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.189 (16)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.249 (16)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e9 families, 9 orders\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 121.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"162\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.133 (18)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.118 (18)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.133 (18)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 414.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"10\" width=\"553\"\u003e\n\u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 12.0pt; font-family: 'Cambria',serif;\"\u003eThe mystery dataset\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 6.5pt; font-family: 'Cambria',serif;\"\u003eCommon features\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 39.55pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.598 (8)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 42.5pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"57\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.370 (18)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: .55in; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"53\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.307 (15)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.615 (8)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.420 (18)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.332 (15)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.659 (8)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.320 (18)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 38.95pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" width=\"52\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.314 (15)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 414.8pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"10\" width=\"553\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cstrong\u003e\u003cspan style=\"font-size: 6.5pt; font-family: 'Cambria',serif;\"\u003eCombination of the common features\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e8 species, 18 families, 15 orders\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 121.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"162\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.272 (41)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.327 (41)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.445 (41)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e8 species, 18 families\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 121.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"162\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.369 (26)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.47 (26)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.415 (26)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e8 species, 15 orders\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 121.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"162\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.297 (23)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.321 (23)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.282 (23)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 59.45pt; border: solid windowtext 1.0pt; border-top: none; padding: 0in 5.4pt 0in 5.4pt;\" width=\"79\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e18 families, 15 orders\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 121.65pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"162\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.255 (33)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.332 (33)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd style=\"width: 116.85pt; border-top: none; border-left: none; border-bottom: solid windowtext 1.0pt; border-right: solid windowtext 1.0pt; padding: 0in 5.4pt 0in 5.4pt;\" colspan=\"3\" width=\"156\"\u003e\n\u003cp style=\"text-align: left;\"\u003e\u003cspan style=\"font-size: 5.0pt; font-family: 'Cambria',serif;\"\u003e0.343 (33)\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"biology-direct","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bdir","sideBox":"Learn more about [Biology Direct](http://biologydirect.biomedcentral.com)","snPcode":"13062","submissionUrl":"https://submission.nature.com/new-submission/13062/3","title":"Biology Direct","twitterHandle":"@Biology_Direct","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Microbiome, OTU, WGS, feature selection, machine learning, Random Forest, Support Vector Machine, Linear Discriminant Analysis, PCoA, ANCOM","lastPublishedDoi":"10.21203/rs.2.20675/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.2.20675/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003c/p\u003e\u003cp\u003eComposition of microbial communities can be location specific, and the different abundance of taxon within location could help us to unravel city-specific signature and predict the sample origin locations accurately. In this study, the whole genome shotgun (WGS) metagenomics data from samples across 16 cities around the world and samples from another 8 cities were provided as the main and mystery datasets respectively as the part of the CAMDA 2019 MetaSUB “Forensic Challenge”. The feature selection, normalization, three methods of machine learning, PCoA (Principal Coordinates Analysis) and ANCOM (Analysis of composition of microbiomes) were conducted for both the main and mystery datasets.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\u003cp\u003eFeature selection, combined with the machines learning methods, revealed that the combination of the common features was effective for predicting the origin of the samples. The average error rates of 11.6% and 30.0% of three machine learning methods were obtained for main and mystery datasets respectively. Using the samples from main dataset to predict the labels of samples from mystery dataset, nearly 89.98% of the test samples could be correctly labeled as “mystery” samples. PCoA showed that nearly 60% of the total variability of the data could be explained by the first two PCoA axes. Although many cities overlapped, the separation of some cities was found in PCoA. The results of ANCOM, combined with importance score from the Random Forest, indicated that the common “family”, “order” of the main-dataset and the common “order” of the mystery dataset provided the most efficient information for prediction respectively.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e\u003c/p\u003e\u003cp\u003eThe results of the classification suggested that the composition of the microbiomes was distinctive across the cities, which was also supported by the results from ANCOM and importance score from the RF. The analysis utilized in this study can be of great help in field of forensic science to efficiently predict the origin of the samples. And the accurate of the prediction could be improved by more samples and better sequencing depth.\u003c/p\u003e","manuscriptTitle":"Unraveling city-specific signature and identifying sample origin locations for the data from CAMDA MetaSUB challenge","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2020-01-13 17:42:23","doi":"10.21203/rs.2.20675/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision","date":"2020-08-17T12:00:00+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2020-08-15T12:00:00+00:00","index":4,"fulltext":"Recommendation: Reviewer's comments unavailable pending editorial decision\n"},{"type":"reviewerAgreed","content":"","date":"2020-07-19T12:00:00+00:00","index":4,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2020-06-26T12:00:00+00:00","index":2,"fulltext":"Recommendation: Endorse publication\nForm responses:\n---\n* Publons Reviewer Recognition. Springer Nature can send verification of this review directly to Publons (a subsidiary of Clarivate Analytics). If you would like to take advantage of this service, please click on the “Yes” option below. Your name, email address, title of the reviewed manuscript, name of the journal, and date of your review submission (the “Review Data”) will then be transmitted to Publons upon publication of the manuscript. If you have already registered at Publons, they will notify you of the receipt of this review and update your profile as per your settings and their policy. If you are not registered with Publons, you will receive an email from them asking you to register in order for them to be able to recognize your review on your new profile page. Publons may use the Review Data to generate derivative metadata for the benefit of Publons and you as a reviewer, carefully considering the sensitivity of such information. For example, Publons may verify your record as a reviewer by updating your profile published on its webservice if you have registered for such service or help editors to identify candidate reviewers. Please find the details of processing in Publons’ privacy policy https://publons.com/about/terms: **Yes**\n* Please indicate how interesting you found the manuscript (whilst Biology Direct does not base publication decisions on interest levels, your feedback is useful for internal purposes):: **\nAn article whose findings are important to those with closely related research interests**\n* Quality of written English: **Needs some language corrections before being published**\n* Declaration of competing interests: **I declare that I have no competing interests.**\n* I agree to Biology Direct’s open peer review policy. I understand that my name will be included on my report to the authors and, if the authors elect to proceed to publication, my named report along with the authors' responses will be included within the published article. I agree for my report to be made available under Open Access Creative Commons CC-BY license . I understand that any comments which I do not wish to be included in my named report can be included as confidential comments to the editors, which will not be published.: **\nI agree to the open peer review policy of the journal**\n* Reviewer summary: **In this paper, the author taxonomically profiled and analyzed the MetaSUB metagenomes from this year’s CAMDA challenge. The authors applied feature selection and machine learning methods to predict the city origin of the samples. I feel it is quite well written and derived some interesting results.**\n* Reviewer recommendations to authors: **Major:\nMy biggest concern is the feature selection applied in this study.\n1) Based on the current text, it seems to me that the authors selected features solely based on their ubiquity across the samples/cities. In machine learning, feature selection aims to find features that best distinguish between classes (in this case, the different cities) in the samples. The procedure described in this manuscript, however, does not serve this purpose. It only selects the OTUs that exist in most samples/cities. These city subway \"housekeeping\" OTUs do not necessarily help to differentiate between cities; the OTUs that are more city-unique, on the other side, were discarded at this stage due to their specificity. I suggest the authors to apply some feature selection methods that aim at finding signature features of each city. For example, Random Forest, one of the three machine learning algorithms used by the authors, performs its own feature selection procedure before the model building. The same measure could, and should, be carried out for the other two methods too.\n2) The common features across the main and mystery sets consists of, not surprisingly, only 5 families and 6 orders. It is thus reasonable to consider them to be common microbes in city subways. However, it is curious why they could, regardless of how universal they are, predict the mystery samples from the main ones at such a high accuracy. It is counter intuitive and worth further investigation. For example, one could add a another PCoA panel including both the main and mystery samples to determine whether there is batch effect.\n3) The authors only considered \"species\", \"family\" and \"order\" levels for feature candidates. It is known that 16S rRNA reaches its maximum resolution at genus level. I suggest the authors to add genus level annotation to the analysis (the authors did mention it in the discussion), as it is likely the most informative and reliable among the others.\n\nMinor:\n1) The authors speculated the sequencing depth to be the main factor that drives the result error rates. A correlation analysis between the two should help to clarify. I also suggest adding information such as sequencing depth and/or sample numbers directly to Figure 1. It is quite doable and will definitely help the readers to understand the results better without digging into supplementary tables.\n2) I feel that this manuscript could benefit a lot from more in-depth discussions. For example, what are the common features? Which species, families and orders? Are they expected to be common in city subways? What are the possible explanations? Or, since some samples in Auckland and Hamilton could not be corrected predicted. Which city/cities were they assigned to by the authors' models? Is there a pattern? Etc.**\n* Minor issues: **1) The font size in the figures needs to be way larger for better readability. \n2) Page 3, line 43, “evidence” instead of “evident”.**\n* Reviewer confidential comments to Editor: ****\n"},{"type":"editorInvitedReview","content":"","date":"2020-06-26T12:00:00+00:00","index":3,"fulltext":"Recommendation: Endorse publication\nForm responses:\n---\n* Publons Reviewer Recognition. Springer Nature can send verification of this review directly to Publons (a subsidiary of Clarivate Analytics). If you would like to take advantage of this service, please click on the “Yes” option below. Your name, email address, title of the reviewed manuscript, name of the journal, and date of your review submission (the “Review Data”) will then be transmitted to Publons upon publication of the manuscript. If you have already registered at Publons, they will notify you of the receipt of this review and update your profile as per your settings and their policy. If you are not registered with Publons, you will receive an email from them asking you to register in order for them to be able to recognize your review on your new profile page. Publons may use the Review Data to generate derivative metadata for the benefit of Publons and you as a reviewer, carefully considering the sensitivity of such information. For example, Publons may verify your record as a reviewer by updating your profile published on its webservice if you have registered for such service or help editors to identify candidate reviewers. Please find the details of processing in Publons’ privacy policy https://publons.com/about/terms: **Yes**\n* Please indicate how interesting you found the manuscript (whilst Biology Direct does not base publication decisions on interest levels, your feedback is useful for internal purposes):: **\nAn article whose findings are important to those with closely related research interests**\n* Quality of written English: **Needs some language corrections before being published**\n* Declaration of competing interests: **I declare that I have no competing interests**\n* I agree to Biology Direct’s open peer review policy. I understand that my name will be included on my report to the authors and, if the authors elect to proceed to publication, my named report along with the authors' responses will be included within the published article. I agree for my report to be made available under Open Access Creative Commons CC-BY license . I understand that any comments which I do not wish to be included in my named report can be included as confidential comments to the editors, which will not be published.: **\nI agree to the open peer review policy of the journal**\n* Reviewer summary: **In this paper, the authors used a number of geo-localised metagenomic samples obtained as part of the CAMDA challenge. They assigned the reads to OTUs, selected the most common OTUs across all samples, and used these as features to build three different models (RF, SVM, LDA).\n\nThey show that the composition of the microbiomes is changing across cities, and that this can be used for classification (and they report low error rates).\n\nOverall, it is an interesting study, though I have a number of specific comments and questions about the approach and the results, detailed below.\n\nThere is a lack of discussion of any similar work, which should be addressed in any revision.\n\nSimilarly, there is not a lot of discussion of the computing environment used in the study. The authors mention a specific cluster, but without giving any information on its specifications, or on the specific resources they were using. There is also a lack of information on the computational performance of the approach. How long does it take to assign OTUs, train, to test?**\n* Reviewer recommendations to authors: **In the introduction, some additional information about the samples could be of interest to the readers (e.g. distributions for the sample size or number of samples per city).\n\nA discussion of the relevant literature also needs to be added.\n\nBecause of the structure of the article, results are described before the methods. To help the readers, it might be useful to give some hints about the methods.\nFor instance, the first sentence of the Results section states that the authors selected the common “species”, “family”, and “order” existing across all the 16 cities, but of course that information was not part of the samples themselves. I would suggest something like \"We used the results of our OTU assignment to select [...]\".\n\nCan the authors provide more details on their cross-validation? They mention randomly selecting one test sample in each run, with 1000 runs repeated. There are 302 samples in the main dataset. Why not systematically leave each sample out in turn?\n\nThe authors need to provide more details on the prediction task on p2 before they introduce Table 3.\nIs the task to take the left-out sample from the dataset, and predict which city it comes from?\nIn particular, for the mystery dataset, are the authors redoing a leave-one-out cross-validation? In other words, are they treating the dataset as if it was another dataset for which they now have labels?\nIf so, one would expect the performance to be much higher than for the main dataset: it is easier to correctly pick one city from 8, than one city from 16. Why are the results largely similar?\n\nThe results on the relationship between sequencing depth and performance (p3) is potentially interesting, but very difficult to follow in its current presentation. It makes it difficult to judge how strong that relationship is. It would be very useful to add a figure that shows the performance as a function of the sequencing depth, for all samples.\nIn that paragraph, it is also not completely clear that \"sequencing depth\" really refers to the sequencing depth. Looking at Tables S1-S3, it seems to be the counts associated with specific OTUs. This is not the same thing as sequencing depth, as some reads have been discarded, and other might just not correspond to these specific OTUs. Can the authors clarify, and update the text (and suggested figure above) accordingly?\n\nThe authors discuss the performance on specific cities, but they do not provide details on the misclassification. For instance, as they mention later, Auckland and Hamilton are in the same country. Are incorrectly classified Auckland samples actually classified as coming from Hamilton? For instance, based on Table S2, AKL_1 looks relatively similar to HAM_8.\nWhat about other cities? Are there common trends in the type of misclassification?\n\nIt is not clear what the last experiment in the \"Machine learning analysis\" section is trying to show (p3-4). From the description, it would seem that the authors took the main dataset (16 cities), and added a \"mystery\" group as a 17th class, made up from 50% of the samples from the mystery dataset. When testing on the remaining 50%, the \"mystery\" samples were corrected assigned to that \"mystery\" class most of the time. Isn't that absolutely expected, given the performance described earlier on the main dataset and on the mystery dataset?\nThe ability of the models to handle unseen cities is of course an important question, but this experiment does not truly address it. What if the models are trained on the main dataset, and then given a sample from the mystery dataset? Are they able to recognise the lack of similarity between that sample and any of the training samples, and if so, can that be used to reliably classify it as a sample from an unknown city?\n\nThe PCoA results (p4) are informative. My main recommendation would be to discuss them at the very start of the results section, as they explain a lot about the performance of the models.\n\nThe microbiome results are interesting but would benefit from a better presentation. For instance, it is not directly clear how this relates with the performance. The x-axis label on Fig.3A is difficult to read. My suggestion is to take this panel, and make it its own figure. If the authors split it into 16 panels (one for each city, e.g. on a 4x4 grid) and show for each city: (i) which abundances are significantly different between that city and all others, and (ii) the overall classification performance for that city, it would really help make the most of those results.\n\nIn the methods section, it would be useful to have more details on the bioinformatics pre-processing. How many reads did the authors start with? How many were excluded?\n\nIn the feature selection section, the authors explain that they are using the most-common OTUs (and then trying to find significant differences in their counts). This is completely valid choice, but it would be interesting to discuss the advantages and limitations of this choice over trying to instead find \"smoking gun\" OTUs that are characteristic of one/few cities.\n\nHave the authors looked at the impact of the number of the trees in the RF?\nFor the SVM, the authors used \"best.svm\" to calculate gamma and c-value. Which samples did they use to perform this? Was this done 1000 times (once for each run), or just once? (same question for LDA)**\n* Minor issues: **The writing should be tidied up (for instance not starting sentences by \"and\").\n\nOn page 3, the authors wrote that \"the samples AKL_1, AKL_7, AKL_14, and HAM_7, HAM_12 cannot be predicted correctly by all three methods\".\nI think that what they meant is that these samples cannot be predicted by any of the three methods.**\n* Reviewer confidential comments to Editor: ****\n"},{"type":"editorInvitedReview","content":"","date":"2020-06-26T12:00:00+00:00","index":1,"fulltext":"Recommendation: Reject as unsound science\nForm responses:\n---\n* Publons Reviewer Recognition. Springer Nature can send verification of this review directly to Publons (a subsidiary of Clarivate Analytics). If you would like to take advantage of this service, please click on the “Yes” option below. Your name, email address, title of the reviewed manuscript, name of the journal, and date of your review submission (the “Review Data”) will then be transmitted to Publons upon publication of the manuscript. If you have already registered at Publons, they will notify you of the receipt of this review and update your profile as per your settings and their policy. If you are not registered with Publons, you will receive an email from them asking you to register in order for them to be able to recognize your review on your new profile page. Publons may use the Review Data to generate derivative metadata for the benefit of Publons and you as a reviewer, carefully considering the sensitivity of such information. For example, Publons may verify your record as a reviewer by updating your profile published on its webservice if you have registered for such service or help editors to identify candidate reviewers. Please find the details of processing in Publons’ privacy policy https://publons.com/about/terms: **Yes**\n* Please indicate how interesting you found the manuscript (whilst Biology Direct does not base publication decisions on interest levels, your feedback is useful for internal purposes):: **\nAn article of limited interest**\n* Quality of written English: **Needs some language corrections before being published**\n* Declaration of competing interests: **I declare that I have no competing interests**\n* I agree to Biology Direct’s open peer review policy. I understand that my name will be included on my report to the authors and, if the authors elect to proceed to publication, my named report along with the authors' responses will be included within the published article. I agree for my report to be made available under Open Access Creative Commons CC-BY license . I understand that any comments which I do not wish to be included in my named report can be included as confidential comments to the editors, which will not be published.: **\nI agree to the open peer review policy of the journal**\n* Reviewer summary: **The authors present an analysis of environmental metagenome sequencing samples provided in context of the CAMDA 2019 competition. In the course of their analysis, different machine learning classifiers (random forests, support vector machines, ...) were trained utilizing and applied on the data sets. In addition, variation of the data was explored using principal coordinates analysis and using compositional analysis of microbes. The work is mostly of descriptive nature and demonstrates that there are differences in microbial signatures between individual cities. However, the claim that the analyses presented in the work help in the field of forensic science is not fully supported. Further extension of the work in the direction would be necessary.\nThe text has a sensible structure but lacks clarity in parts, which makes it difficult to retrace the steps of analysis at times. For instance the Methods section could be more detailed and parts outlined in the Results section should actually go into Methods.\nIn general, the manuscript summarizes some interesting first observations, but needs more work in order to support the made claims.**\n* Reviewer recommendations to authors: **I have the following major points of criticism / questions:\n\n* Both in the Abstract and in the Discussion the authors make the claim of application in forensics, e.g.: \"... indicating that the combination of the common features could be the effective microbial fingerprint for unraveling city-specific signature and identifying sample origin locations, which proved to be with great potential for forensic science.\" The experiments have shown that some of the city samples can be assigned to the correct city label if a labeled training set is available. However, this is an overly optimistic assumption for forensics. When looking into the mystery samples, the only prediction that is made is whether a sample has the \"mystery\" label or not. This only contains very little information for forensic applications. From the data and experiments at hand it cannot be judged, how well a sample of unknown origin could be classified.\n* When describing their strategy of feature selection, the text lacks sufficient clarity. It would be helpful to see an example of what \"combined feature\" means. For instance, the authors write: \"The combinations of the common “species”, “family”, and “order” were regarded as the combined features.\" What does \"combined mean in this context? Also it is unclear how the rules for feature selection were derived and what the reasoning is behind them. Lastly, the whole section of feature selection in Results has a more procedural character and could be fused with the similarly titled part in the Methods section.\n* When measuring performance of their machine learning models, the authors use leave-one-out cross validation. The entire available data is used as this validation set and no held-out independent test set is used for performance measurement. Hence, the reported performance is likely overly optimistic. Additionally, the cross validation has been run 1000 times. However, the number of total samples is only 302. As the splits of data were randomly selected, some constellations were used more often than others. How was the number 1000 determined?\n* The authors note in their discussion that \"some cities such as London and Oslo were separated from most cities, indicating the unique composition of microbiomes in these cities\". However, earlier in the text the authors note that \"Upon finding excessive zeros in samples from London, the samples of London could be easily identified from all the samples\". This makes it very likely that the high discriminative power towards London samples actually stems from a technical artifact rather than from a biological signal. Similarly, the Oslo samples might well be just different on a technical scale (e.g. read length, paired-end state, etc) and not have a specific microbial fingerprint per se. All these aspects are currently left out of consideration and lead to potential mis-interpretation of the results.\n* Many of the differences between samples can possibly be ascribed to technical factors, such as low sequencing depth, unbalanced sample set distribution, read error rates, etc. However, the only measure of normalization used and mentioned in the text is a log2-cpm normalization. What other options for normalization are there? In addition, the section on quality control lacks many details. For instance, it is mentioned that \"samples with poor sequencing depth were removed\", but not how the threshold was selected and what \"poor\" actually means.\n* In the presentation of the principal coordinates analysis, the authors observe that Auckland and Hamilton overlap and make the link to the fact that both cities are in New Zealand. However, this overlap does not seem to be of any significance as many other cities also overlap (some of them even more compactly). So what can be learned from this statement?\n\nIn addition to the points above, I would like to raise the following minor points:\n* The authors write that they implemented several machine learning methods. From the text it seams that mostly pre-written packages were used for analysis. It would be helpful if any code were provided, so the analyses could be recapitulated.\n* The authors write that \"features selection was implemented [...] to help us obtain the appropriate features for prediction.\" What does \"appropriate\" mean in this context?\n* In the PCoA, the authors describe the amount of variation explained by the first two axes. How is to be interpreted in the presence of potential confounders and technical variability?\n* The text describing the machine learning methods for SVM states: \"...the SVM model with the appropriate parameters was obtained by testing the performance of models with different parameters.\" What dataset is this model selection based on?**\n* Minor issues: **In some sentences, the language is not clear enough and it is unclear what is meant.\n* Methods: \"each method was implemented 1000 times\" --\u003e does this mean the method was run 1000 times?\n* Methods: \"Bray-Cruits distance\" --\u003e does this refer to Bray-Curtis dissimilarity? (it does not meet all the criteria for a distance)\n* Spelling / Grammar needs to be checked throughout the text.**\n* Reviewer confidential comments to Editor: **would need a major revision**\n"},{"type":"reviewerAgreed","content":"","date":"2020-06-10T12:00:00+00:00","index":3,"fulltext":""},{"type":"reviewerAgreed","content":"","date":"2020-06-08T12:00:00+00:00","index":2,"fulltext":""},{"type":"reviewerAgreed","content":"","date":"2020-03-02T12:00:00+00:00","index":1,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2020-02-07T12:00:00+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2020-01-21T12:00:00+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2020-01-20T12:00:00+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2020-01-10T12:00:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"","date":"2020-01-09T12:00:00+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"biology-direct","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bdir","sideBox":"Learn more about [Biology Direct](http://biologydirect.biomedcentral.com)","snPcode":"13062","submissionUrl":"https://submission.nature.com/new-submission/13062/3","title":"Biology Direct","twitterHandle":"@Biology_Direct","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"19ff05b2-973f-4801-8866-23b2beebaf07","owner":[],"postedDate":"January 13th, 2020","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":49078,"name":"General Microbiology"}],"tags":[],"updatedAt":"2021-01-10T15:06:11+00:00","versionOfRecord":{"articleIdentity":"rs-11276","link":"https://doi.org/10.1186/s13062-020-00284-1","journal":{"identity":"biology-direct","isVorOnly":false,"title":"Biology Direct"},"publishedOn":"2021-01-04 15:03:57","publishedOnDateReadable":"January 4th, 2021"},"versionCreatedAt":"2020-01-13 17:42:23","video":"","vorDoi":"10.1186/s13062-020-00284-1","vorDoiUrl":"https://doi.org/10.1186/s13062-020-00284-1","workflowStages":[]},"version":"v1","identity":"rs-11276","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"identity":"rs-11276","version":["v1"]},"buildId":"ApUGefWb6u5IBVtyqm6d5","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}