The Clinical Value of Alpha-diversity metrics to establish Dysbiosis in Microbiome Studies

The Clinical Value of Alpha-diversity metrics to establish Dysbiosis in Microbiome Studies | 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 Article The Clinical Value of Alpha-diversity metrics to establish Dysbiosis in Microbiome Studies Nataša Mortvanski, Anna Sayol-Altarriba, Andrea Aira, Elisa Rubio, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7987343/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Background Selecting appropriate alpha diversity metrics is essential for capturing the biological and clinical relevance of gut microbiome studies. However, no consensus or clear rationale currently guides this choice. In this study, we compared the distribution of 10 commonly used alpha diversity metrics in patients with inflammatory bowel disease (IBD) and Clostridioides difficile infection (CDI) to those in a healthy reference population. The healthy reference group consisted of fecal donor samples validated for clinical use. Our aim was to benchmark the performance of these metrics against a healthy gut microbiome. Results We found that improvements in health status were associated with increases in both richness and evenness components of alpha diversity. The ability to differentiate between health conditions varied among metrics. Notably, the Gini index emerged as a robust evenness metric for predicting health status and detecting group differences, while most richness metrics showed consistent trends across all comparisons. Conclusions These results suggest that alpha diversity metrics can serve as valuable tools to capture microbiome disturbances and monitoring gut microbiome health. Health sciences/Biomarkers Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Health sciences/Gastroenterology Health sciences/Health care Health sciences/Medical research Biological sciences/Microbiology Gut microbiota dysbiosis microbial alpha diversity Clostridioides difficile infection (CDI) inflammatory bowel disease (IBD) fecal microbiota transplantation (FMT) short-chain fatty acids (SCFAs) Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 INTRODUCTION The human gut microbial community, also known as the gut microbiome, plays a critical role in maintaining host health and well-being. The gut microbiota is tightly connected to the immune system, primarily through mechanisms like the intestinal epithelial barrier, which defends against pathogen colonization and invasion (Carding et al., 2015 ) but also indirectly, through the production of short-chain fatty acids (SCFA) that may drive immune cells towards a more immunotolerant phenotype (Kim, 2023 ). Disruptions in this microbial ecosystem, referred to as intestinal dysbiosis, are often characterized by a loss of diversity and reduced community functionality (Mosca et al., 2016 ). The administration of broad spectrum antibiotics or sudden dietary changes (e.g. reduction of fiber intake) are the major sources of exogenous dysbiosis (Gill et al., 2021 ; Lathakumari et al., 2024 ). Also, dysbiosis has been linked to several diseases, including inflammatory bowel disease (IBD), diabetes, multiple sclerosis, celiac disease, and colorectal cancer (Carding et al., 2015 ; Lupp et al., 2007 ; Mosca et al., 2016 ; Turnbaugh et al., 2009 ). However, the definition and study of dysbiosis remains a major challenge. One of the key and somewhat unexpected findings of the Human Microbiome Project was that there is no universal “core healthy microbiome” (Huse et al., 2012). Microbiome composition varies greatly between individuals, and overlapping microbial functions suggest that health is not tied to a single microbial profile. Instead, multiple configurations of beneficial microbes can support health, forming a spectrum of healthy states (Manor et al., 2020 ). This complexity challenges traditional views of dysbiosis and prompts a rethinking of what defines a “healthy” microbiome. Alpha diversity metrics are commonly used to summarize microbiome diversity by estimating their richness (number of taxonomic groups) and their evenness (distribution of group abundances). Numerous studies have explored alpha diversity as a potential biomarker for health conditions (Hagerty et al., 2020 ; Khanna et al., 2016 ; Li et al., 2022 ; Manor et al., 2020 ; Plassais et al., 2021 ). However, concerns remain about the selection and benchmarking of these metrics, as well as their statistical validity (Kers & Saccenti, 2022 ). Importantly, as noted by Johnson & Burnet ( 2016 ), higher diversity does not always equate to a healthier or more stable microbiome. Diversity metrics often fail to capture species composition and their interactions, which are critical for assessing community quality. The gut microbial communities must be understood as a complex ecosystem where microbes interact with each other and their environment, influencing host health in various ways (Belkaid & Hand, 2014 ; Culp & Goodman, 2023 ). One key functional read-out of this ecosystem is the production of SCFAs, such as acetate, propionate, and butyrate. SCFAs are primarily generated when gut bacteria ferment dietary fiber, highlighting the importance of fiber-rich diets in maintaining a healthy microbiome (Gill et al., 2021 ; Koh et al., 2016 ). These metabolites play a crucial role in supporting host health by serving as an energy source for intestinal epithelial cells, strengthening the gut barrier, and regulating inflammation (Alva-Murillo et al., 2012 ; Koh et al., 2016 ; Suzuki et al., 2008 ; Yao et al., 2022 ). Higher levels of SCFAs are often linked to a diverse, well-balanced microbiome and are associated with improved metabolic and immune function (Lange et al., 2023; Yao et al., 2022 ). The study of SCFA production offers valuable insights into the microbiome’s functional capacity, providing a more comprehensive view of ecosystem health than taxonomy alone. To address these knowledge gaps, we systematically compared alpha diversity metrics using patient data from well-curated sources, including the American Gut Project (AGP) and a healthy reference cohort from the stool biobank at Hospital Clínic de Barcelona (HCB). Our analysis focused on two common conditions: inflammatory bowel disease (IBD) and Clostridioides difficile infection (CDI). IBD encompasses two major subtypes: Crohn’s disease (CD) and ulcerative colitis (UC); whereas CDI is a well-characterized infection that responds effectively to fecal microbiota transplantation (FMT). Both conditions are consistently associated with reduced alpha diversity (Abdel-Rahman & Morgan, 2023; Clooney et al., 2021; Van Werkhoven et al., 2021). METHODS Data selection We queried the Qiita repository (Gonzalez et al., 2018 ) for suitable datasets. Using this platform helped minimized technical differences between studies, as all datasets were processed uniformly. Additionally, Qiita provides standardized sample metadata for each study. The American Gut Project (AGP) dataset (McDonald et al., 2018 ) is currently the largest available dataset on Qiita, with participants from around the world, primarily from North America and Europe (UK). An extensive metadata file, containing 700 features, was also included. We defined our healthy population as individuals aged 20 to 69 years, with a normal BMI (18.5 to 25), who reported not taking antibiotics in the past year nor having been diagnosed with IBD, irritable bowel syndrome (IBS), or CDI. This selection process yielded 1470 samples, which formed our “healthy” population. Furthermore, five studies on IBD and CDI (Khanna et al., 2017 ; Lloyd-Price et al., 2019 ; Vázquez-Baeza et al., 2018 ; Webside: Qiita ID 11549; Weingarden et al., 2015 ) from the Qiita repository were included in the analysis. All datasets obtained from Qiita were processed according to the 16S rRNA gene sequencing protocols established by the Earth Microbiome Project (Thompson et al., 2017 ) and the Human Microbiome Project (HMP Consortium, 2012 ). Studies were selected based on similarities in sequencing platform, data processing steps, and the specific 16S rRNA region sequenced (Supplementary Table S1 ). Only studies analyzing human adult fecal microbiome samples were included. Due to the lack of appropriate CDI studies on Qiita, we added the Khanna et al. 2016 dataset from the BioProject repository (Supplementary Table S2). Additionally, a dataset generated at Hospital Clínic de Barcelona (HCB) was used, which includes samples from CDI patients, recipients of FMT, and stool donors from the stool bank (Aira et al., 2022 ) (Supplementary Table S3). Raw sequencing data generated in HCB have been deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession PRJNA1337461. All patients were comprehensively screened for pathogens following current guidelines. All experiments involving human fecal samples were performed in accordance with relevant guidelines and regulations. The study protocol was approved by the Ethics Committee of Research with medicines (CEIm) of HCB (Ref HCB/2016/0824 and Ref HCB/2024/0239), with appropriate informed consent from all participants in compliance with the Declaration of Helsinki and applicable national and European legislation. Data processing All Qiita datasets were analysed by Qiita’s Quantitative Insights Into Microbial Ecology (QIIME 2, version 2022.2.1) analysis plugin (Bolyen et al., 2019 ). BioProject and HCB data were processed using QIIME 2 pipeline (version 2020.8.0). Quality control was performed using Deblur (Amir et al., 2017 ) for the BioProject CDI dataset and the DADA2 plugin (Callahan et al., 2016 ) for HCB datasets (Supplementary section 1.1). After trimming we were left with reads with over 200 bp length and a median quality score above 30. The rarefaction depths for each dataset were chosen based on the respective rarefaction curves (Supplementary Table S1 ). Multiple sequence alignment was done using MAFFT (Katoh, 2002 ) and a phylogenetic tree was created using FastTree (Price et al., 2009 ). Taxonomy was assigned using a Pre-fitted sklearn-based taxonomy classifier (Bokulich et al., 2018 ) trained with the Greengenes 13_8 16S rRNA gene database (McDonald et al., 2012 ). Ten commonly used alpha diversity metrics, available as outputs from the standard QIIME2 pipeline, were calculated: Chao1, Margalef, Menhinick, Fisher alpha, Faith's PD, Gini index, Strong’s dominance, Pielou’s evenness, Shannon’s entropy, and Simpson’s index. SCFAs quantification of stool donor samples Stool donor samples of the stool bank (n = 102) of HCB were assessed for SCFAs quantification. The quantification of acetate, propionate and butyrate was performed using a previously described method (Sayol-Altarriba et al., 2024 ). Briefly, acidified supernatants were centrifuged, doubly extracted, and derivatized. After this processing, samples were quantified by gas chromatography coupled with mass spectrometry (GC/MS) and the resulting SCFAs levels were normalized using the bacterial count of the original stool sample assessed by flow cytometry (UF-400, Sysmex Co, Kobe, Japan), Statistical analysis A detailed study design that includes the different datasets used and the analysis performed can be found in Supplementary section 1.3. The analysis of the obtained alpha diversity data was done in R (version 4.1.2) and can be divided into three parts: Characterizing the distributions of alpha metrics in the AGP healthy population. Comparing alpha diversity of AGP healthy population to IBD and CDI populations. Confirming observed trends by comparing HCB data from healthy stool donors and CDI patients. Populations of interest were compared visually by plotting the difference of population distributions and the significance of the difference was estimated by applying a non-parametric Mann-Whitney-Wilcoxon test (with significance level p < 0.05). We used this non-parametric test since none of the diversity metrics can be considered as normally distributed. A Random Forest classifier (randomForest package) was used to quantify the importance of different metrics for predicting sample’s health status. First the selected data was separated into training (70% of the data) and test (30% of the data). Then, the groups of healthy and unhealthy samples in the training set were balanced (undersampling) so that we have the same number of healthy and unhealthy samples. We trained 26 different random forest classifiers. One of the models included all alpha diversity metrics as predictor variables, while the other 25 consisted of all possible combinations of one "richness" and one "evenness" alpha metric (including Shannon’s entropy and Simpson’s index). We analyzed which models showed the best accuracy. Furthermore, we applied the function wilcox_effsize from the package rstatix to compute the effect size of health condition on each of the diversity indices. The Wilcoxon effect size (r) was calculated as the Z statistic divided by the square root of the sample size (N). The function also produced confidence intervals using bootstrapping. We also tested the applicability of Crawford & Howell's (1998) modified t-test for single-case studies in the context of estimating the abnormality of an individual sample’s alpha diversity value in comparison to a defined control population. Furthermore, this method was proven to have modest inflation of Type I error in skewed control populations, unlike traditionally used z scores (Crawford & Garthwaite, 2005 ), thus being even more appropriate for our analysis. The formula for the test is as follows: $$\:{t}_{n-1}^{}=\frac{{x}^{}-\underset{\_}{x}}{s{\sqrt{\frac{n+1}{n}}}_{}^{}}$$ where x* is the patient’s score, \(\:\underset{\_}{x}\) and S are the mean and SD of scores in the control sample, and n is the size of the control sample. For the first experiment we defined the healthy (control) population as 70% of all healthy AGP samples (n = 1033), while the other 30% together with CDI samples (Khanna et al., 2016 ) form a “test” population. Each test sample was compared to the control population using the modified t-test. The probability of each sample belonging to the control population was calculated, representing the percentage of samples from the control population that have lower value than a tested sample (Crawford & Garthwaite, 2005 ). The second experiment consisted in comparing CDI samples before and after FMT, as well as the stool donors’ samples used for FMT, with control population of stool donor samples (n = 113) from HCB. Finally, the metabolic composition of stool donor’s samples from HCB (n = 102) was analyzed to assess the correlation between the SCFAs and the abundance of bacterial taxa and alpha diversity indexes. For this, the Spearman’s method was applied between pairs of variables, and the p- values were adjusted for multiple testing using the Benjamini-Hochberg correction ( q -values). Data and code availability All used datasets are publicly available at Qiita and NCBI BioProject repository, except from HCB sequences and metadata. Code produced for this analysis, alongside with raw and processed data files is available in a public repository (see data availability section below). RESULTS Characterization of AGP population’s alpha diversity The alpha diversity metrics obtained from the healthy samples of the AGP dataset (n = 1470) were used to describe the relationship between different metrics and to explore non-clinical factors affecting diversity. Table 1 contains the definitions and similarities between chosen alpha diversity metrics. In the literature these metrics are classified as estimations of richness, evenness or a combination of both. As expected, Pearson correlation proved that many of these indices are highly correlated (Figure 1a). Gini and Strong’s dominance indexes approach the estimation of alpha diversity in a reciprocal way compared to the other metrics (the higher the index, the lower the diversity/evenness). When inverted, both Gini and Strong’s index are closer to the group of indices that account for both richness and evenness (Shannon entropy and Simpsons’s index) or just evenness (Pielou’s evenness) component of diversity. By applying Exploratory Factor Analysis (psych package in R) we confirmed that there are two underlying latent factors (Supplementary Fig. S1), likely richness and evenness, that are driving this grouping of indices. Furthermore, the distributions of alpha metrics were tested for normality using the Shapiro–Wilk test. The skewness and kurtosis were also computed for all metrics (Supplementary Table S4). Although none of the metrics in our analysis satisfied the criteria for normality, richness-related metrics showed closer resemblance to a normal distribution (Figure 1b). Table 1. Definitions of alpha diversity indices used in this work Metric Definition References Richness metrics Chao 1 Estimator of species richness, including rare or unobserved species (Chao, 1984) Margalef’s index Ration of number of species in the community relative to the total number of individuals (Magurran, 2004) Menhinick’s index Ratio of number of species to square root of number of individuals in the sample (Magurran, 2004) Fisher’s alpha Measuring the relationship between the number of species and the relative abundance of each species (Fisher et al., 1943) Faith’s phylogenetic diversity The sum of OUT branch lengths. It takes into account phylogenetic distance between OUTs (Faith, 1992) Evenness metrics Gini index Measure of evenness or inequality in the distribution of microbial species within a community. High Gini index – high inequality in the distribution of species (due to one highly dominant species or a few dominant species with many rare species) (Gini, 1912) Strong’s dominance index Abundance unevenness or dominance concentration. High SDI specifically indicates that the most abundant species is highly dominant, even if there are other species present in smaller quantities (Strong, 2002) Pielou evenness Measure of relative evenness of species richness. Compares the observed diversity to the maximum possible diversity – normalized version of Shannon’s index (Pielou, 1966) Both richness and evenness Shannon’s entropy Heterogeneity of a sample, redundancy, entropy (Shannon & Weaver, 1949) Simpson’s index Probability that any two organisms sampled will be the sample phylotype (Simpson, 1949) More detailed explanations of each metric with their mathematical formulas can be found in Supplementary section 1.2. To determine the non-clinical factors affecting alpha diversity within the healthy population, first we compared numerical features (age, height, weight, and BMI) with alpha diversity metrics and found no significant correlation. For categorical variables we attempted to estimate feature importance using a Random Forest classifier (Supplementary Fig. S2). The most important categories in both approaches were country of residence and country of birth, suggesting that geographically different populations might differ in alpha diversity. Further investigation confirmed that participants from European countries (both by birth and residence) had a higher average richness and evenness than those from North American countries. Additionally, participants of Asian origin (country of birth) had the lowest diversity and significantly differed from European samples in most metrics (Supplementary Fig. S3 and S4). Alpha diversity comparison between control and case populations One of the aims of this analysis was to assess whether and to what extent we can capture the differences in microbial communities between populations of healthy controls and samples of patients with diagnosed IBD or CDI, by only looking at their alpha diversity. Difference in alpha diversity between IBD samples and control Alpha diversity values obtained from two studies related to IBD (Lloyd-Price et al., 2019; Webside: Qiita ID 11549) were compared to those calculated for the AGP control group. Two IBD conditions, Crohn’s disease - CD (n = 26) and ulcerative colitis - UC (n = 40) showed lower richness, but higher evenness when compared with the control samples. The exceptions are Faith’s PD (significant difference with healthy samples; p < 0.05) and Menhinick (not significant with p = 0.1) based on which CD had the highest richness (Supplementary Fig. S5). Furthermore, UC showed significant differences from the control population in more alpha metrics than CD (Supplementary Tables S5 and S6). This can be due to the heterogeneous severity of Crohn’s disease diagnosis or a low number of available samples. However, information about the severity was not provided in metadata. To increase the sample size and gain further insights, data from a longitudinal CD study (Vázquez-Baeza et al., 2018) were added to the analysis (n = 293). Control samples from healthy family members were included in this dataset (n = 353). This control population had a significantly different distribution mean from the AGP control population for most of the alpha metrics (Supplementary Table S7), as did the CD samples from previous IBD datasets (Supplementary Table S8). Therefore, we proceeded with the analysis with the original study groups. We confirmed the following conclusions from the original study: a) Crohn’s disease samples have significantly lower richness and evenness compared to controls (Supplementary Table S9); b) CD samples that previously underwent some surgical intervention had lower alpha diversity than those that did not (Supplementary Table S10, Supplementary Fig. S6). This difference was significant based on the Mann-Whitney-Wilcoxon test for more alpha metrics than those in the first CD dataset. Difference in alpha diversity between CDI samples and control and effect of FMT on alpha diversity Alpha diversity of CDI samples (n = 73) from Khanna et al. (2016) showed highly significant differences from AGP healthy samples, showing lower richness and lower evenness (Supplementary Table S11, Supplementary Fig. S7). This condition is the only one treated with FMT. Studies examining the success of this type of treatment usually estimate it by comparing the composition of the microbial community before and after treatment. However, we wanted to quantify the improvement of the microbial community after FMT through alpha diversity. For this we examined a longitudinal study of 4 CDI patients treated with FMT whose progress was examined in multiple time points after procedure yielding 88 post-FMT samples (Weingarden et al., 2015). We noticed a consistent trend of increased microbial richness and evenness over time after the procedure in all subjects (Figure 2). Although the post-FMT diversity values varied over time, all time points after FMT showed higher richness and lower dominance than the initial value before treatment. Furthermore, we examined whether the same trend of worse response of CDI patients to FMT in cases with an underlying IBD suggested by Khanna et al. (2017) could be observed on the alpha diversity level. Supplementary Fig. S9 shows the trend of improvement in alpha diversity on days 7th and 28th day after FMT. For all alpha indices, the diversity increases toward the donor's mean value. However, samples with underlying CD showed a richness and evenness decrease on the 28th day compared to 7th day for all alpha metrics (except for Pielou index). Difference in alpha diversity between CDI and healthy donor samples from Hospital Clínic de Barcelona We compared alpha diversity in CDI samples before (n = 18) and after transplantation (n = 38), as well as the difference between CDI patients and healthy stool donors (n = 151). The results show a significant difference between CDI samples before FMT and stool donor samples for all alpha metrics (Figure 3). Furthermore, FMT increased samples’ richness (significantly for all richness metrics except form Faith PD) and evenness (lower Gini and Strong, higher Shannon, Pielou, and Simpson although for the last two the difference was not significant; Supplementary Table S12). Additionally, we also noticed that donor samples from the HCB have higher richness (except from Menhinick) and much higher evenness than AGP samples that we have previously used as controls (Figure 4). Statistical power analysis Kers & Saccenti (2022) highlighted the difference in the statistical power of various alpha metrics when comparing the two groups. Wilcoxon effect size was used to determine which metric from each of the groups (richness and evenness) is the most sensitive to the difference between healthy and unhealthy samples. Table 2. Definitions of alpha diversity indices used in this work Parameter p -value adjusted a Effect size (r) Confidence low Confidence high Magnitude Gini index (x) <10 -16 0.553 0.53 0.58 Large Chao1 (+) <10 -16 0.400 0.36 0.43 Moderate Menhinick <10 -16 0.377 0.34 0.41 Moderate Margalef (+) <10 -16 0.375 0.34 0.41 Moderate Fisher alpha (+) <10 -16 0.375 0.34 0.41 Moderate Faith PD (+) <10 -16 0.302 0.26 0.35 Moderate Shannon entropy (*) <10 -16 0.249 0.22 0.29 Small Simpson (*) <10 -16 0.189 0.15 0.22 Small Pielou evenness (x) 1,19e -10 0.135 0.10 0.17 Small Strong (x) 0.0022 0.064 0.02 0.11 Small a p -value adjusted is a result of the Mann-Whitney-Wilcoxon test for the difference between means of all healthy samples (n = 1823) and unhealthy sample (n = 432). Adjustment is done using the false discovery rate (FDR) method. Wilcoxon effect size (r) with estimated confidence interval varies from 0 to 1 and is commonly interpreted as: 0.10 – 0.3 (small effect), 0.30 – 0.5 (moderate effect) and >= 0.5 (large effect). (+) richness metric, (x) evenness metric, (*) both richness and evenness. Based on the results shown in Table 2, the Gini index seems to show the highest effect size when comparing healthy and unhealthy samples, scoring better than the other metrics in the “evenness” group. Chao1 is the second best, and the best among the “richness” metrics. Feature importance and accuracy of Random Forest classifier We compared the classifier's performance on different datasets to determine if it was more sensitive to one disease than another. Additionally, we investigated whether the classifier was better at predicting the overall health status (healthy or unhealthy) of the samples or predicting specific conditions. The highest accuracy was obtained with a model containing all alpha metrics when it was trained on all datasets together (~88% for predicting condition, ~89% for health status; Supplementary Table S13), and in the case of IBD and the healthy dataset alone (~87% for predicting condition, ~88% for health status; Supplementary Table S14). However, the decrease in accuracy is not substantial when we choose only two metrics (one representing the richness metric and the other the evenness metric). The highest accuracy was achieved when the model used the Gini index as an evenness metric (82-85% for predicting condition, 84–86% for health status). Even higher accuracy was obtained when the random forest classifier was trained on the healthy and CDI datasets using different models (Supplementary Table S15). This classifier appears to be able to determine the differences between healthy and CDI samples with more than 90% accuracy in all the models, with each model containing Gini index reaching 100% accuracy (except for the Gini-Faith PD model with 99.8% accuracy). Similar results were obtained from the classification of HCB CDI and healthy donor samples (Supplementary Table S16). In this case, more models showed absolute accuracy, but we also had fewer samples for classification in this dataset. Again, models including the Gini index scored the best, with the addition of two models: Menhinick-Strong and Menhinick-Shannon, that exhibited the same accuracy. For all classifiers, models with the Gini index performed the best. Therefore, the Gini index scored the highest importance for Random Forest classifier (Supplementary Fig. S10). The best “richness” metric varied between different classifiers, although the accuracy was not affected much when the “evenness” metric was Gini index. Modified t-test for single case comparisons The probability of each test population sample belonging to the control (AGP) population from the first experiment is summarized in Table 3. On average, healthy samples exhibited a higher probability of belonging to the control population compared to CDI samples. “Richness” metrics showed great results with all CDI samples located in the 5% (Faith PD) to 12% (Fisher alpha) of the control population. In contrast, “evenness” metrics showed significant overlap between CDI and healthy test samples, with the exception of the Gini index, which achieved complete separation. These findings indicate that "evenness" metrics, when analyzed using t-tests, may not provide sufficient discriminatory capability to distinguish between healthy and unhealthy samples. Table 3. p -values obtained by modified t-test: AGP as control population Diversity index Probability [Mean (Min, Max)] N healthy overlap CDI* CDI (n = 73) Healthy (n = 437) Chao1 0.026 (0.005, 0.082) 0.477 (0.013, 1.00) 27 (6%) Margalef 0.030 (0.004, 0.116) 0.481 (0.008, 1.00) 54 (12%) Menhinick 0.025 (0.004, 0.092) 0.481 (0.008, 1.00) 40 (9%) Fisher alpha 0.043 (0.014, 0.122) 0.474 (0.020, 1.00) 53 (12%) Faith PD 0.008 (0.001, 0.024) 0.482 (0.005, 0.999) 5 (1%) Gini index 0 (0, 0) 0.531 (0.0002, 0.948) 0 (0%) Strong 0.480 (0.016, 0.973) 0.494 (0.0024, 0.999) 410 (94%) Pielou evenness 0.405 (0.010, 0.887) 0.520 (0.0001, 0.915) 429 (98%) Shannon entropy 0.180 (0.003, 0.628) 0.511 (0.0004, 0.965) 246 (56%) Simpson 0.371 (0.001, 0.728) 0.536 (0, 0.79) 315 (72%) p -value mean, maximum and minimum were calculated from the results of applying modified t-test on “test” population. T value was calculated based on Crawford & Howell (1998) formula. The obtained value was compared to t-distribution with one tailed t-test observing lower end. *How many healthy samples have lower probability of belonging to the control population than the CDI sample with the highest probability. Based on Table 4, there is an increase in the probability of FMT recipients and donor samples belonging to the HCB donor population compared to CDI samples before FMT. The “richness” metrics, along with the Gini index, demonstrated better performance in distinguishing between healthy and unhealthy samples. Among these metrics, the Chao1 index exhibited the least overlap between donor samples and CDI before FMT, while showing the greatest overlap with CDI after FMT- an outcome aligned with expectations. The Gini index also provided good discriminatory power, achieving complete separation for donor samples and CDI samples before FMT. Table 4. p- values obtained by modified t-test: stool donors from HCB as control population Diversity index Probability [Mean (Min, Max)] CDI pre overlap ‡ CDI post overlap ‡ CDI pre (n = 18) CDI post (n = 38) Donors (n = 38) Chao1 0.032 (0, 0.218) 0.206 (0, 0.995) 0.562 (0.018, 1) 2 (5%) 26 (68%) Margalef 0.044 (0, 0.328) 0.220 (0, 0.983) 0.574 (0.016, 1) 2 (5%) 12 (32%) Menhinic k 0.242 (0, 0.948) 0.523 (0.001, 1) 0.616 (0.016, 1) 2 (5%) 12 (32%) Fisher alpha 0.062 (0, 0.430) 0.263 (0.001, 0.998) 0.586 (0.021, 1) 2 (5%) 21 (55%) Gini index* 0 (0, 0) 0.0003 (0, 0.003) 0.027 (0, 0.488) 0 (0%) 19 (50%) Strong* 0.091 (0, 0.860) 0.090 (0, 0.888) 0.181 (0, 0.679) 38 (100%) 38 (100%) Pielou evenness 0.078 (0, 0.650) 0.083 (0, 0.628) 0.486 (0.003, 0.949) 32 (84%) 35 (92%) Shannon entropy 0.011 (0, 0.095) 0.089 (0, 0.798) 0.535 (0.001, 1) 2 (5%) 20 (53%) Simpson 0.041 (0, 0.395) 0.103 (0, 0.822) 0.569 (0, 0.983) 4 (10%) 21 (55%) *For Gini index and Strong dominance probability is calculated for the upper tail of T distribution since healthy samples have lower values. ‡ How many donor samples have lower probability of belonging the control population than the CDI sample before and after FMT with the highest probability. Desirable outcomes would be lower overlap with CDI pre and higher overlap with CDI post. SCFAs correlation with bacterial genera and alpha-diversity indexes Results of the metabolic composition (SCFAs) of stool donors’ samples from HCB were correlated with the abundance of bacterial taxa and alpha diversity indexes. Of the 78 bacterial genera detected in the metagenomic study, 8 showed correlation with one of the main SCFAs ( q -value < 0.01). Higher SCFAs levels in stool were associated with bacterial genera responsible for producing SCFAs, such as Anaerostipes , Bacteroides , Coprococcus , Faecalibacterium , Odoribacter , and Roseburia . The AF12 genera, not known for its capacity to produce SCFAs, also presented a positive correlation with acetate, although very weak. On the other hand, Oxalobacter inversely correlated with fecal butyrate (Figure 5a). Regarding alpha diversity indexes, SCFAs showed a significant inverse correlation with 8 out of the 10 alpha diversity indexes assessed ( q -value < 0.01), being fecal butyrate levels the ones that present more significant correlations. Only Gini’s index (which measures the inequality in the distribution of microbial species) shows a positive correlation with butyrate. Strong’s dominance index showed no significant correlation with any of SCFAs analyzed (Figure 5b). DISCUSSION Gut microbiome diversity is increasingly recognized as a key indicator of intestinal health. However, a consistent definition of dysbiosis remains elusive due to its conceptual ambiguity. Benchmarking alpha diversity metrics against well-characterized healthy and diseased cohorts offers a valuable strategy for assessing their clinical relevance. Such benchmarks provide evidence that these metrics can reliably reflect microbial disturbances. In particular, richness and evenness indices show strong alignment with gut health, supporting their utility in both clinical and research contexts. Nonetheless, this study underscores important limitations in interpreting alpha diversity metrics, especially when short-chain fatty acids (SCFAs) are used as indirect proxies for microbial diversity. Our findings demonstrate that healthy gut microbial communities are consistently characterized by higher richness and evenness compared to unhealthy states such as inflammatory bowel disease (IBD) or C. difficile infection (CDI). Notably, CDI samples displayed robust and significant reductions in richness, with recovery observed post-fecal microbiota transplant (FMT), reinforcing the role of alpha diversity metrics in tracking intervention efficacy. However, the higher richness observed in the Lloyd-Price et al. ( 2019 ) CDI samples, compared to AGP controls highlights the challenges posed by disease heterogeneity and the importance of using well-characterized control groups. c While alpha diversity is a powerful summary measure, it is important to acknowledge its limitations. Metrics such as Chao1 and Faith's PD quantify richness but fail to capture microbial composition and functional interactions. By contrast, the Gini index, an underutilized evenness metric, emerged as a robust discriminator between healthy and unhealthy samples across datasets. Combining richness and evenness metrics, as demonstrated by our predictive models, significantly improved accuracy, particularly in distinguishing severe dysbiosis such as CDI. SCFAs, including butyrate and acetate, are often viewed as beneficial markers of microbial functionality. However, our results highlight a counterintuitive inverse correlation between SCFA levels and alpha diversity metrics. This aligns with prior studies (De La Cuesta-Zuluaga et al., 2018 ; Jones et al., 2021 ), suggesting that SCFA abundance reflects the activity of specific taxa (e.g., Faecalibacterium, Anaerostipes, Roseburia ), rather than the overall diversity of the ecosystem. Furthermore, factors such as SCFA absorption in the colon (90–95%), daily food choices, transit time, and the time of stool collection introduce variability, undermining stool SCFA concentrations as a direct indicator of microbial production or diversity (Johnson et al., 2019 ; Jones et al., 2021 ; McNeil et al., 1978 ; Vogt & Wolever, 2003 ). These findings emphasize that while SCFAs play a critical role in gut health, they cannot serve as standalone markers for microbial diversity. Alpha diversity metrics, which consider all taxa, provide a broader snapshot of ecosystem structure, albeit without functional specificity. Several limitations warrant consideration. The absence of high-quality, publicly available control datasets with comprehensive clinical metadata limits the generalizability of our findings. For instance, the AGP dataset, while extensive, relies on self-reported health statuses, which may not accurately reflect microbiome profiles. Conversely, well-characterized donor populations, such as those from HCB, are limited in size but hold promise for large-scale validation studies. Geographical variability introduces additional confounding factors, with differences in lifestyle and diet influencing diversity metrics. This underscores the need for population-specific controls when interpreting alpha diversity benchmarks. Despite the challenges in defining dysbiosis, our study demonstrates that alpha diversity metrics, particularly combinations of richness and evenness indices, are valuable markers of gut microbiota health. Metrics like the Gini index, Chao1, and Faith’s PD provide robust, complementary insights into microbial diversity, while SCFA abundance, although functionally significant and informative, remains an imperfect proxy of microbial diversity. Addressing current limitations, including dataset quality and population variability, will be essential for advancing the clinical application of these metrics in diagnosing and monitoring gut microbiome disturbances. Declarations Ethics approval and consent to participate The analysis of samples from HCB patients was approved by the Ethics Committee of Research with medicines (CEIm) of HCB (Ref HCB/2016/0824 and Ref HCB/2024/0239), with appropriate informed consent from all participants in compliance with the Declaration of Helsinki and applicable national and European legislation. Consent for publication Not applicable. Funding This study has recived support from the grant CEX2023-0001290-S funded by MCIN/AEI/ 10.13039/501100011033, and support from the Generalitat de Catalunya through the CERCA Program. ASA was supported by the PREDOCS-UB 2022 grant from the University of Barcelona (UB). Availability of data and materials Data for the analyses in this study included datasets from public repositories (see the Methods section) and data from patients at HCB. Raw sequencing reads from HCB patients are available in NCBI Sequence Read Archive (SRA) under BioProject accession PRJNA1337461, with Biosample accessions SAMN52298702-SAMN52298908, and SRA run accessions SRR35728543-SRR35728749. All data are publicly accessible at: https://www.ncbi.nlm.nih.gov/sra/PRJNA1337461 All the alpha-diversity data and code required to reproduce the analysis presented in this paper, including links to the original repositories, is available in the following GitHub repository: https://github.com/CDB-coreBM/mortvanski_et_al_2025. Acknowledgments We acknowledge the Departments of Microbiology and Biochemistry and Molecular Genetics for their contribution. References Aira, A., Rubio, E., Ruiz, A., Vergara, A., Casals-Pascual, C., Rico, V., Suñé-Negre, J.M., & Soriano, A. (2022). New Procedure to Maintain Fecal Microbiota in a Dry Matrix Ready to Encapsulate. Front. Cell. Infect. Microbiol. 12 , 899257. Alva-Murillo, N., Ochoa-Zarzosa, A., & López-Meza, J.E. (2012). Short chain fatty acids (propionic and hexanoic) decrease Staphylococcus aureus internalization into bovine mammary epithelial cells and modulate antimicrobial peptide expression. Vet. Microbiol. 155 , 324–331. Amir, A., McDonald, D., Navas-Molina, J.A., Kopylova, E., Morton, J.T., Zech Xu, Z., Kightley, E.P., Thompson, L.R., Hyde, E.R., Gonzalez, A., & Knight, R. (2017). Deblur Rapidly Resolves Single-Nucleotide Community Sequence Patterns. mSystems 2 , e00191-16. Belkaid, Y., & Hand, T.W. (2014). Role of the Microbiota in Immunity and Inflammation. Cell 157 , 121–141. Bokulich, N.A., Kaehler, B.D., Rideout, J.R., Dillon, M., Bolyen, E., Knight, R., Huttley, G.A., & Gregory Caporaso, J. (2018). Optimizing taxonomic classification of marker-gene amplicon sequences with QIIME 2’s q2-feature-classifier plugin. Microbiome 6 , 90. Bolyen, E., Rideout, J.R., Dillon, M.R., et al. (2019). Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 37 , 852–857. Callahan, B.J., McMurdie, P.J., Rosen, M.J., Han, A.W., Johnson, A.J.A., & Holmes, S.P. (2016). DADA2: High resolution sample inference from Illumina amplicon data. Nat. Methods 13 , 581–583. Carding, S., Verbeke, K., Vipond, D.T., Corfe, B.M., & Owen, L.J. (2015). Dysbiosis of the gut microbiota in disease. Microb. Ecol. Health Dis. 26 . Chao, A. (1984). Nonparametric Estimation of the Number of Classes in a Population. Scand. J. Stat. 11 , 265–270. Crawford, J.R., & Garthwaite, P.H. (2005). Testing for Suspected Impairments and Dissociations in Single-Case Studies in Neuropsychology: Evaluation of Alternatives Using Monte Carlo Simulations and Revised Tests for Dissociations. Neuropsychology 19 , 318–331. Crawford, J.R., & Howell, D.C. (1998). Comparing an Individual’s Test Score Against Norms Derived from Small Samples. Clin. Neuropsychol. 12 , 482–486. Culp, E.J., & Goodman, A.L. (2023). Cross-feeding in the gut microbiome: Ecology and mechanisms. Cell Host Microbe 31 , 485–499. De La Cuesta-Zuluaga, J., Mueller, N., Álvarez-Quintero, R., Velásquez-Mejía, E., Sierra, J., Corrales-Agudelo, V., Carmona, J., Abad, J., & Escobar, J. (2018). Higher Fecal Short-Chain Fatty Acid Levels Are Associated with Gut Microbiome Dysbiosis, Obesity, Hypertension and Cardiometabolic Disease Risk Factors. Nutrients 11 , 51. Faith, D.P. (1992). Conservation evaluation and phylogenetic diversity. Biol. Conserv. 61 , 1–10. Fisher, R.A., Corbet, A.S., & Williams, C.B. (1943). The Relation Between the Number of Species and the Number of Individuals in a Random Sample of an Animal Population. J. Anim. Ecol. 12 , 42. Gill, S.K., Rossi, M., Bajka, B., & Whelan, K. (2021). Dietary fibre in gastrointestinal health and disease. Nat. Rev. Gastroenterol. Hepatol. 18 , 101–116. Gini, C. (1912). Variabilita e Mutabilita contributo allo studio delle distribuzioni e delle relazioni statistiche. (Bologna: Tipografia di Paolo Cuppini). Gonzalez, A., Navas-Molina, J.A., Kosciolek, T., et al. (2018). Qiita: rapid, web-enabled microbiome meta-analysis. Nat. Methods 15 , 796–798. Hagerty, S.L., Hutchison, K.E., Lowry, C.A., & Bryan, A.D. (2020). An empirically derived method for measuring human gut microbiome alpha diversity: Demonstrated utility in predicting health-related outcomes among a human clinical sample. PLOS ONE 15 , e0229204. HMP Consortium (2012). Structure, Function and Diversity of the Healthy Human Microbiome. Nature 486 , 207–214. Johnson, K.V.-A., & Burnet, P.W.J. (2016). Microbiome: Should we diversify from diversity? Gut Microbes 7 , 455–458. Johnson, A.J., Vangay, P., Al-Ghalith, G.A., Hillmann, B.M., Ward, T.L., Shields-Cutler, R.R., Kim, A.D., Shmagel, A.K., Syed, A.N., Walter, J., Menon, R., Koecher, K., & Knights, D. (2019). Daily Sampling Reveals Personalized Diet-Microbiome Associations in Humans. Cell Host Microbe 25 , 789-802.e5. Jones, J., Reinke, S.N., Ali, A., Palmer, D.J., & Christophersen, C.T. (2021). Fecal sample collection methods and time of day impact microbiome composition and short chain fatty acid concentrations. Sci. Rep. 11 , 13964. Katoh, K. (2002). MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30 , 3059–3066. Kers, J.G., & Saccenti, E. (2022). The Power of Microbiome Studies: Some Considerations on Which Alpha and Beta Metrics to Use and How to Report Results. Front. Microbiol. 12 , 796025. Khanna, S., Montassier, E., Schmidt, B., Patel, R., Knights, D., Pardi, D.S., & Kashyap, P.C. (2016). Gut microbiome predictors of treatment response and recurrence in primary Clostridium difficile infection. Aliment. Pharmacol. Ther. 44 , 715–727. Khanna, S., Vazquez-Baeza, Y., González, A., et al. (2017). Changes in microbial ecology after fecal microbiota transplantation for recurrent C. difficile infection affected by underlying inflammatory bowel disease. Microbiome 5 , 55. Kim, C.H. (2023). Complex regulatory effects of gut microbial short-chain fatty acids on immune tolerance and autoimmunity. Cell. Mol. Immunol. 20 , 341–350. Koh, A., De Vadder, F., Kovatcheva-Datchary, P., & Bäckhed, F. (2016). From Dietary Fiber to Host Physiology: Short-Chain Fatty Acids as Key Bacterial Metabolites. Cell 165 , 1332–1345. Lathakumari, R.H., Vajravelu, L.K., Satheesan, A., Ravi, S., & Thulukanam, J. (2024). Antibiotics and the gut microbiome: Understanding the impact on human health. Med. Microecol. 20 , 100106. Li, Z., Zhou, J., Liang, H., Ye, L., Lan, L., Lu, F., Wang, Q., Lei, T., Yang, X., Cui, P., & Huang, J. (2022). Differences in Alpha Diversity of Gut Microbiota in Neurological Diseases. Front. Neurosci. 16 , 879318. Lloyd-Price, J., Arze, C., Ananthakrishnan, A.N., et al. (2019). Multi-omics of the gut microbial ecosystem in inflammatory bowel diseases. Nature 569 , 655–662. Lupp, C., Robertson, M.L., Wickham, M.E., Sekirov, I., Champion, O.L., Gaynor, E.C., & Finlay, B.B. (2007). Host-Mediated Inflammation Disrupts the Intestinal Microbiota and Promotes the Overgrowth of Enterobacteriaceae. Cell Host Microbe 2 , 119–129. Magurran, A.E. (2004). Measuring Biological Diversity. Blackwell . Manor, O., Dai, C.L., Kornilov, S.A., Smith, B., Price, N.D., Lovejoy, J.C., Gibbons, S.M., & Magis, A.T. (2020). Health and disease markers correlate with gut microbiome composition across thousands of people. Nat. Commun. 11 , 5206. McDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A., Andersen, G.L., Knight, R., & Hugenholtz, P. (2012). An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. ISME J. 6 , 610–618. McDonald, D., Hyde, E., Debelius, J.W., et al. (2018). American Gut: an Open Platform for Citizen Science Microbiome Research. 3 . McNeil, N.I., Cummings, J.H., & James, W.P. (1978). Short chain fatty acid absorption by the human large intestine. Gut 19 , 819–822. Mosca, A., Leclerc, M., & Hugot, J.P. (2016). Gut Microbiota Diversity and Human Diseases: Should We Reintroduce Key Predators in Our Ecosystem? Front. Microbiol. 7 . Pielou, E.C. (1966). The Measurement of Diversity in Different Types of Biological Colledions. J. Theor. Biol. 13 , 131–144. Plassais, J., Gbikpi-Benissan, G., Figarol, M., Scheperjans, F., Gorochov, G., Derkinderen, P., & Cervino, A.C.L. (2021). Gut microbiome alpha-diversity is not a marker of Parkinson’s disease and multiple sclerosis. Brain Commun. 3 , fcab113. Price, M.N., Dehal, P.S., & Arkin, A.P. (2009). FastTree: Computing Large Minimum Evolution Trees with Profiles instead of a Distance Matrix. Mol. Biol. Evol. 26 , 1641–1650. Sayol-Altarriba, A., Aira, A., Villasante, A., Albarracín, R., Faneca, J., Casals, G., Villanueva-Cañas, J.L., & Casals-Pascual, C. (2024). Normalization of short-chain fatty acid concentration by bacterial count of stool samples improves discrimination between eubiotic and dysbiotic gut microbiota caused by Clostridioides difficile infection-. Gut Microbes 16 , 2415488. Shannon, C.E., & Weaver, W. (1949). The Mathematical Theory of Communication (Urbana, IL: The University of Illinois Press). Simpson, E. (1949). Measurement of Diversity. 163 , 688. Strong, W.L. (2002). Assessing species abundance unevenness within and between plant communities. Community Ecol. 3 , 237–246. Suzuki, T., Yoshida, S., & Hara, H. (2008). Physiological concentrations of short-chain fatty acids immediately suppress colonic epithelial permeability. Br. J. Nutr. 100 , 297–305. Thompson, L.R., Sanders, J.G., McDonald, D., et al. (2017). A communal catalogue reveals Earth’s multiscale microbial diversity. Nature 551 , 457–463. Turnbaugh, P.J., Hamady, M., Yatsunenko, T., et al. (2009). A core gut microbiome in obese and lean twins. Nature 457 , 480–484. Vázquez-Baeza, Y., Gonzalez, A., Xu, Z.Z., Washburne, A., Herfarth, H.H., Sartor, R.B., & Knight, R. (2018). Guiding longitudinal sampling in IBD cohorts. Gut 67 , 1743–1745. Vogt, J.A., & Wolever, T.M.S. (2003). Fecal Acetate Is Inversely Related to Acetate Absorption from the Human Rectum and Distal Colon. J. Nutr. 133 , 3145–3148. Webside: Qiita ID 11549 Qiita ID 11549. Metaomics Reveals Microbiome Based Proteolysis as a Driver of Ulcerative Colitis Severity. Weingarden, A., González, A., Vázquez-Baeza, Y., Weiss, S., Humphry, G., Berg-Lyons, D., Knights, D., Unno, T., Bobr, A., Kang, J., Khoruts, A., Knight, R., & Sadowsky, M.J. (2015). Dynamic changes in short- and long-term bacterial composition following fecal microbiota transplantation for recurrent Clostridium difficile infection. Microbiome 3 , 10. Yao, L., Davidson, E.A., Shaikh, M.W., Forsyth, C.B., Prenni, J.E., & Broeckling, C.D. (2022). Quantitative analysis of short-chain fatty acids in human plasma and serum by GC–MS. Anal. Bioanal. Chem. 414 , 4391–4399. Additional Declarations No competing interests reported. Supplementary Files Supplementarymaterialalphadiversity.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 11 Nov, 2025 Editor assigned by journal 11 Nov, 2025 Editor invited by journal 10 Nov, 2025 Submission checks completed at journal 06 Nov, 2025 First submitted to journal 06 Nov, 2025 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-7987343","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":546122180,"identity":"5407219e-296b-4f50-9b4e-838bff4ae3d1","order_by":0,"name":"Nataša Mortvanski","email":"","orcid":"","institution":"ISGlobal","correspondingAuthor":false,"prefix":"","firstName":"Nataša","middleName":"","lastName":"Mortvanski","suffix":""},{"id":546122181,"identity":"3729368e-f2a6-4369-b4bd-f131330f3f5f","order_by":1,"name":"Anna Sayol-Altarriba","email":"","orcid":"","institution":"Univeristy of Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Sayol-Altarriba","suffix":""},{"id":546122182,"identity":"06399624-584f-4caf-824b-822cde0176db","order_by":2,"name":"Andrea Aira","email":"","orcid":"","institution":"Hospital Clínic de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Andrea","middleName":"","lastName":"Aira","suffix":""},{"id":546122183,"identity":"19583d2b-85c2-4d08-bc04-ec7abb6e32f2","order_by":3,"name":"Elisa Rubio","email":"","orcid":"","institution":"Hospital Clínic de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Elisa","middleName":"","lastName":"Rubio","suffix":""},{"id":546122184,"identity":"46b15a1b-cde2-428f-b6c3-1648681a6f8f","order_by":4,"name":"Gregori Casals","email":"","orcid":"","institution":"Hospital Clínic de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Gregori","middleName":"","lastName":"Casals","suffix":""},{"id":546122185,"identity":"5e48b41e-9efb-493c-b03a-b214c9f5dfff","order_by":5,"name":"Alex Soriano","email":"","orcid":"","institution":"Hospital Clínic de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Alex","middleName":"","lastName":"Soriano","suffix":""},{"id":546122186,"identity":"5bc85597-5bcb-4b78-8bd5-4a9f14866169","order_by":6,"name":"José Luis Villanueva-Cañas","email":"","orcid":"","institution":"Hospital Clínic de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"José","middleName":"Luis","lastName":"Villanueva-Cañas","suffix":""},{"id":546122187,"identity":"9fa386f7-9dc5-4faa-9d0a-7a2c5fd2fae8","order_by":7,"name":"Climent Casals-Pascual","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxUlEQVRIiWNgGAWjYNCCAgY5BnbGBoYE4rUYMBgzMMO0sBGpJbGBGcYhpIW//fDjDz8M6tL7m5nbHjyoYZDnn9+AX4vEmTQzyR6Dw7kzDjO2GyQcYzCccYyQkyQYzBh4DA7kbmBmbJNIBHmHsBb2zx//AB1mANMiT1gLj4E0jwFzAlyLASEtEmdyyqRlDA4bAv3SJpFwTMJw47EE/Fr4249v/vimok6ev739meSPGht5ucMHCFiDbitpykfBKBgFo2AUYAcAbyE6Ugn6blkAAAAASUVORK5CYII=","orcid":"","institution":"Hospital Clínic de Barcelona","correspondingAuthor":true,"prefix":"","firstName":"Climent","middleName":"","lastName":"Casals-Pascual","suffix":""}],"badges":[],"createdAt":"2025-10-30 09:08:32","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7987343/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7987343/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96492863,"identity":"ae737e52-760a-464c-9360-990fc6ef74a7","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":963960,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptalphadiversityV2revised.docx","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/46e3122be5a9d776f8bb5c90.docx"},{"id":96492859,"identity":"50c81893-ecf2-43ee-9cce-fafdb8e4b1e9","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9465,"visible":true,"origin":"","legend":"","description":"","filename":"160b99a64adb4d018b55885eb21b09f8.json","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/131f31993949ee83c73274ae.json"},{"id":96604046,"identity":"111a2efa-38fa-401a-96b0-bd7b6f31a8fc","added_by":"auto","created_at":"2025-11-24 09:12:32","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1287932,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterialalphadiversity.docx","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/04836408d54712ad613f070a.docx"},{"id":96492877,"identity":"0cddaa5c-ee42-4d88-a710-2e689876c26c","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":138136,"visible":true,"origin":"","legend":"","description":"","filename":"160b99a64adb4d018b55885eb21b09f81enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/487747730e53b55832df7199.xml"},{"id":96492866,"identity":"ae379f4f-771a-4bfa-92bb-02d2172f8885","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":98560,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/511e093edcf066d7c3515a7b.png"},{"id":96492868,"identity":"eb4033c3-08d4-4f58-84cd-1a8e8073dd8f","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":125267,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/7710bd8f42f0a04449d2d95d.png"},{"id":96603021,"identity":"fe5c4651-1b54-4e27-a9f4-288bac0a0a79","added_by":"auto","created_at":"2025-11-24 09:06:06","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":313824,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/a58ba02d0920a90c7203887a.png"},{"id":96492870,"identity":"9af2756b-6bcb-4a5e-8b90-f69b4226069d","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":110228,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/14e9ad30f5c8544b85a27aee.png"},{"id":96603031,"identity":"9cbb692d-7af8-4b9b-a0a7-374bbdd51fd8","added_by":"auto","created_at":"2025-11-24 09:06:13","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118568,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/f3770960fb94203403b2bd27.png"},{"id":96604146,"identity":"5b5cb9e0-da2e-4efa-abfb-5e1ccbf1a396","added_by":"auto","created_at":"2025-11-24 09:12:56","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":194959,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/1bbb5c9b57e312966ff949f9.jpeg"},{"id":96604415,"identity":"a6986b3c-05dc-46fa-b160-8486d138c43f","added_by":"auto","created_at":"2025-11-24 09:13:53","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24866,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/c20ada8c069f0cc94db18503.png"},{"id":96604117,"identity":"955743ee-417f-48fc-8ac3-50b25d88c89f","added_by":"auto","created_at":"2025-11-24 09:12:48","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22054,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/ea75a8a5cc490236ad4b146a.png"},{"id":96492874,"identity":"b4dd6efe-7a8c-41ef-a156-143de4dd3a95","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56585,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/0bbf8d964f0a9c1e6c28e821.png"},{"id":96603779,"identity":"570a6f1b-1038-4367-b790-2b5c76461aa2","added_by":"auto","created_at":"2025-11-24 09:11:30","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21532,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/d5e8ad29ccd6f50fe514cd98.png"},{"id":96492878,"identity":"d3875355-cba0-4f91-9b24-30bdd9612440","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22991,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/ff16de316d5e4d6c35672a0b.png"},{"id":96492879,"identity":"d13dc02e-aed1-4f82-b1c6-b7136ea6cd1d","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39236,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/4e96af7ee7c4a973fa02120e.png"},{"id":96492881,"identity":"3c65ccc2-db70-4fc5-8d09-c410b6b2fb27","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"xml","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":135410,"visible":true,"origin":"","legend":"","description":"","filename":"160b99a64adb4d018b55885eb21b09f81structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/29258ae83c06e1a1d91a85fa.xml"},{"id":96492880,"identity":"e7ddfa7f-e61c-44e1-b64b-936a8b7c7ade","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"html","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":146573,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/ce37c286fc273c21a059daee.html"},{"id":96603777,"identity":"ba28b875-8d36-4e6a-b077-7b2e90667962","added_by":"auto","created_at":"2025-11-24 09:11:30","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":281400,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(a) \u003c/strong\u003eCorrelation plot for 10 alpha diversity metrics obtained from the healthy population sampled from the American Gut Project dataset. Based on hierarchical clustering, there are three groups of correlated metrics: one consisting of indices connected to richness, second with metrics accounting for evenness, and Gini and Strong making the third (evenness) group. \u003cstrong\u003e(b)\u003c/strong\u003e Diversity distribution shows different ranges of different alpha diversity metrics. Distribution of metrics accounting for evenness are more skewed than the ones accounting for richness.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/ff0f409b4bf20372a023e25b.png"},{"id":96492860,"identity":"d033b23a-6203-4904-b741-08581be83913","added_by":"auto","created_at":"2025-11-21 18:12:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":313824,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProgression of alpha diversity metrics over time after FMT.\u003c/strong\u003e Samples before FMT were taken at different timepoints before FMT (15, 1 or 0 days before FMT) but are represented at the same position for clarity. The day of FMT is represented by a dashed line. For progressions of all 10 used alpha metrics check the Supplementary Fig. S8. (+) richness metric, (x) evenness metric, (*) both richness and evenness.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/7ae232ee3a647f2c3bd4b39a.png"},{"id":96604307,"identity":"4ef370f7-164e-42b5-a88f-a62bb3118195","added_by":"auto","created_at":"2025-11-24 09:13:32","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":110228,"visible":true,"origin":"","legend":"\u003cp\u003eKolmogrov-Smimov test was applied to calculate the significance of the difference in population means of different conditions. CDIpre = CDI samples before FMT treatment (n = 18), CDIpost = CDI samples after FMT treatment (n = 38), donor = stool donor samples (n = 151). (+) richness metrics, (x) evenness metrics, (*) both richness and evenness.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/c0dbe86767e1fad3c2117f12.png"},{"id":96603902,"identity":"b373c008-bd40-4834-9772-3736f6d4e442","added_by":"auto","created_at":"2025-11-24 09:12:01","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":118568,"visible":true,"origin":"","legend":"\u003cp\u003eRichness metrics show an expected progression from more severe CDI dataset (Khanna et al., 2016), Hospital’s CDI samples before and after FMT treatment and healthy samples (AGP and stool donor samples). Evenness on the other hand is unexpectedly low for AGP samples (all “evenness” metrics) and high for CDI (Strong). (+) richness metrics, (x) evenness metrics, (*) both richness and evenness.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/b280f2baae304a1239fba76b.png"},{"id":96603062,"identity":"6b6974f6-a7c0-46a3-8bb0-aab0477c63cf","added_by":"auto","created_at":"2025-11-24 09:06:32","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":114850,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation matrix between SCFAs concentrations of stool samples normalized by the bacterial count and:\u003cstrong\u003e (a)\u003c/strong\u003e bacterial genus; \u003cstrong\u003e(b) \u003c/strong\u003ealpha diversity indexes according to Spearman’s rho. Only significant correlations (q-value \u0026lt; 0.01) are shown in the matrix\u003c/p\u003e","description":"","filename":"5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/e97ec6a3b730bc0a5333177b.jpeg"},{"id":96708353,"identity":"15c96e37-de3c-4940-a712-4ca926c10714","added_by":"auto","created_at":"2025-11-25 10:01:20","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2002804,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/c8aabdd7-f16d-4fb3-90ff-bb8fd74d9fd0.pdf"},{"id":96603931,"identity":"ba5bafef-fb0a-41e7-8da8-9223f749316f","added_by":"auto","created_at":"2025-11-24 09:12:05","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1287932,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterialalphadiversity.docx","url":"https://assets-eu.researchsquare.com/files/rs-7987343/v1/191d058ca943bf8ea14bb2f5.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"The Clinical Value of Alpha-diversity metrics to establish Dysbiosis in Microbiome Studies","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eThe human gut microbial community, also known as the gut microbiome, plays a critical role in maintaining host health and well-being. The gut microbiota is tightly connected to the immune system, primarily through mechanisms like the intestinal epithelial barrier, which defends against pathogen colonization and invasion (Carding et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) but also indirectly, through the production of short-chain fatty acids (SCFA) that may drive immune cells towards a more immunotolerant phenotype (Kim, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Disruptions in this microbial ecosystem, referred to as intestinal dysbiosis, are often characterized by a loss of diversity and reduced community functionality (Mosca et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The administration of broad spectrum antibiotics or sudden dietary changes (e.g. reduction of fiber intake) are the major sources of exogenous dysbiosis (Gill et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Lathakumari et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Also, dysbiosis has been linked to several diseases, including inflammatory bowel disease (IBD), diabetes, multiple sclerosis, celiac disease, and colorectal cancer (Carding et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Lupp et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Mosca et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Turnbaugh et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). However, the definition and study of dysbiosis remains a major challenge.\u003c/p\u003e\u003cp\u003eOne of the key and somewhat unexpected findings of the Human Microbiome Project was that there is no universal \u0026ldquo;core healthy microbiome\u0026rdquo; (Huse et al., 2012). Microbiome composition varies greatly between individuals, and overlapping microbial functions suggest that health is not tied to a single microbial profile. Instead, multiple configurations of beneficial microbes can support health, forming a spectrum of healthy states (Manor et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This complexity challenges traditional views of dysbiosis and prompts a rethinking of what defines a \u0026ldquo;healthy\u0026rdquo; microbiome.\u003c/p\u003e\u003cp\u003eAlpha diversity metrics are commonly used to summarize microbiome diversity by estimating their richness (number of taxonomic groups) and their evenness (distribution of group abundances). Numerous studies have explored alpha diversity as a potential biomarker for health conditions (Hagerty et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Khanna et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Manor et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Plassais et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). However, concerns remain about the selection and benchmarking of these metrics, as well as their statistical validity (Kers \u0026amp; Saccenti, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Importantly, as noted by Johnson \u0026amp; Burnet (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), higher diversity does not always equate to a healthier or more stable microbiome. Diversity metrics often fail to capture species composition and their interactions, which are critical for assessing community quality.\u003c/p\u003e\u003cp\u003eThe gut microbial communities must be understood as a complex ecosystem where microbes interact with each other and their environment, influencing host health in various ways (Belkaid \u0026amp; Hand, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Culp \u0026amp; Goodman, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). One key functional read-out of this ecosystem is the production of SCFAs, such as acetate, propionate, and butyrate. SCFAs are primarily generated when gut bacteria ferment dietary fiber, highlighting the importance of fiber-rich diets in maintaining a healthy microbiome (Gill et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Koh et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). These metabolites play a crucial role in supporting host health by serving as an energy source for intestinal epithelial cells, strengthening the gut barrier, and regulating inflammation (Alva-Murillo et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Koh et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Suzuki et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Yao et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Higher levels of SCFAs are often linked to a diverse, well-balanced microbiome and are associated with improved metabolic and immune function (Lange et al., 2023; Yao et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The study of SCFA production offers valuable insights into the microbiome\u0026rsquo;s functional capacity, providing a more comprehensive view of ecosystem health than taxonomy alone.\u003c/p\u003e\u003cp\u003eTo address these knowledge gaps, we systematically compared alpha diversity metrics using patient data from well-curated sources, including the American Gut Project (AGP) and a healthy reference cohort from the stool biobank at Hospital Cl\u0026iacute;nic de Barcelona (HCB). Our analysis focused on two common conditions: inflammatory bowel disease (IBD) and \u003cem\u003eClostridioides difficile\u003c/em\u003e infection (CDI). IBD encompasses two major subtypes: Crohn\u0026rsquo;s disease (CD) and ulcerative colitis (UC); whereas CDI is a well-characterized infection that responds effectively to fecal microbiota transplantation (FMT). Both conditions are consistently associated with reduced alpha diversity (Abdel-Rahman \u0026amp; Morgan, 2023; Clooney et al., 2021; Van Werkhoven et al., 2021).\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eData selection\u003c/h2\u003e\u003cp\u003eWe queried the Qiita repository (Gonzalez et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) for suitable datasets. Using this platform helped minimized technical differences between studies, as all datasets were processed uniformly. Additionally, Qiita provides standardized sample metadata for each study.\u003c/p\u003e\u003cp\u003eThe American Gut Project (AGP) dataset (McDonald et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) is currently the largest available dataset on Qiita, with participants from around the world, primarily from North America and Europe (UK). An extensive metadata file, containing 700 features, was also included. We defined our healthy population as individuals aged 20 to 69 years, with a normal BMI (18.5 to 25), who reported not taking antibiotics in the past year nor having been diagnosed with IBD, irritable bowel syndrome (IBS), or CDI. This selection process yielded 1470 samples, which formed our \u0026ldquo;healthy\u0026rdquo; population.\u003c/p\u003e\u003cp\u003eFurthermore, five studies on IBD and CDI (Khanna et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Lloyd-Price et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; V\u0026aacute;zquez-Baeza et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Webside: Qiita ID 11549; Weingarden et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) from the Qiita repository were included in the analysis. All datasets obtained from Qiita were processed according to the 16S rRNA gene sequencing protocols established by the Earth Microbiome Project (Thompson et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) and the Human Microbiome Project (HMP Consortium, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Studies were selected based on similarities in sequencing platform, data processing steps, and the specific 16S rRNA region sequenced (Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Only studies analyzing human adult fecal microbiome samples were included.\u003c/p\u003e\u003cp\u003eDue to the lack of appropriate CDI studies on Qiita, we added the Khanna et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2016\u003c/span\u003e dataset from the BioProject repository (Supplementary Table S2). Additionally, a dataset generated at Hospital Cl\u0026iacute;nic de Barcelona (HCB) was used, which includes samples from CDI patients, recipients of FMT, and stool donors from the stool bank (Aira et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) (Supplementary Table S3). Raw sequencing data generated in HCB have been deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession PRJNA1337461. All patients were comprehensively screened for pathogens following current guidelines.\u003c/p\u003e\u003cp\u003e All experiments involving human fecal samples were performed in accordance with relevant guidelines and regulations. The study protocol was approved by the Ethics Committee of Research with medicines (CEIm) of HCB (Ref HCB/2016/0824 and Ref HCB/2024/0239), with appropriate informed consent from all participants in compliance with the Declaration of Helsinki and applicable national and European legislation.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eData processing\u003c/h3\u003e\n\u003cp\u003eAll Qiita datasets were analysed by Qiita\u0026rsquo;s Quantitative Insights Into Microbial Ecology (QIIME 2, version 2022.2.1) analysis plugin (Bolyen et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). BioProject and HCB data were processed using QIIME 2 pipeline (version 2020.8.0). Quality control was performed using Deblur (Amir et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) for the BioProject CDI dataset and the DADA2 plugin (Callahan et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) for HCB datasets (Supplementary section 1.1). After trimming we were left with reads with over 200 bp length and a median quality score above 30. The rarefaction depths for each dataset were chosen based on the respective rarefaction curves (Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Multiple sequence alignment was done using MAFFT (Katoh, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2002\u003c/span\u003e) and a phylogenetic tree was created using FastTree (Price et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). Taxonomy was assigned using a Pre-fitted sklearn-based taxonomy classifier (Bokulich et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) trained with the Greengenes 13_8 16S rRNA gene database (McDonald et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Ten commonly used alpha diversity metrics, available as outputs from the standard QIIME2 pipeline, were calculated: Chao1, Margalef, Menhinick, Fisher alpha, Faith's PD, Gini index, Strong\u0026rsquo;s dominance, Pielou\u0026rsquo;s evenness, Shannon\u0026rsquo;s entropy, and Simpson\u0026rsquo;s index.\u003c/p\u003e\n\u003ch3\u003eSCFAs quantification of stool donor samples\u003c/h3\u003e\n\u003cp\u003eStool donor samples of the stool bank (n\u0026thinsp;=\u0026thinsp;102) of HCB were assessed for SCFAs quantification. The quantification of acetate, propionate and butyrate was performed using a previously described method (Sayol-Altarriba et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Briefly, acidified supernatants were centrifuged, doubly extracted, and derivatized. After this processing, samples were quantified by gas chromatography coupled with mass spectrometry (GC/MS) and the resulting SCFAs levels were normalized using the bacterial count of the original stool sample assessed by flow cytometry (UF-400, Sysmex Co, Kobe, Japan),\u003c/p\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eA detailed study design that includes the different datasets used and the analysis performed can be found in Supplementary section 1.3. The analysis of the obtained alpha diversity data was done in R (version 4.1.2) and can be divided into three parts:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eCharacterizing the distributions of alpha metrics in the AGP healthy population.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eComparing alpha diversity of AGP healthy population to IBD and CDI populations.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eConfirming observed trends by comparing HCB data from healthy stool donors and CDI patients.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003ePopulations of interest were compared visually by plotting the difference of population distributions and the significance of the difference was estimated by applying a non-parametric Mann-Whitney-Wilcoxon test (with significance level p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). We used this non-parametric test since none of the diversity metrics can be considered as normally distributed.\u003c/p\u003e\u003cp\u003eA Random Forest classifier (randomForest package) was used to quantify the importance of different metrics for predicting sample\u0026rsquo;s health status. First the selected data was separated into training (70% of the data) and test (30% of the data). Then, the groups of healthy and unhealthy samples in the training set were balanced (undersampling) so that we have the same number of healthy and unhealthy samples. We trained 26 different random forest classifiers. One of the models included all alpha diversity metrics as predictor variables, while the other 25 consisted of all possible combinations of one \"richness\" and one \"evenness\" alpha metric (including Shannon\u0026rsquo;s entropy and Simpson\u0026rsquo;s index). We analyzed which models showed the best accuracy.\u003c/p\u003e\u003cp\u003eFurthermore, we applied the function wilcox_effsize from the package rstatix to compute the effect size of health condition on each of the diversity indices. The Wilcoxon effect size (r) was calculated as the Z statistic divided by the square root of the sample size (N). The function also produced confidence intervals using bootstrapping.\u003c/p\u003e\u003cp\u003eWe also tested the applicability of Crawford \u0026amp; Howell's (1998) modified t-test for single-case studies in the context of estimating the abnormality of an individual sample\u0026rsquo;s alpha diversity value in comparison to a defined control population. Furthermore, this method was proven to have modest inflation of Type I error in skewed control populations, unlike traditionally used z scores (Crawford \u0026amp; Garthwaite, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), thus being even more appropriate for our analysis. The formula for the test is as follows:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{t}_{n-1}^{}=\\frac{{x}^{}-\\underset{\\_}{x}}{s{\\sqrt{\\frac{n+1}{n}}}_{}^{}}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003ewhere x* is the patient\u0026rsquo;s score, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\underset{\\_}{x}\\)\u003c/span\u003e\u003c/span\u003e and S are the mean and SD of scores in the control sample, and n is the size of the control sample. For the first experiment we defined the healthy (control) population as 70% of all healthy AGP samples (n\u0026thinsp;=\u0026thinsp;1033), while the other 30% together with CDI samples (Khanna et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) form a \u0026ldquo;test\u0026rdquo; population. Each test sample was compared to the control population using the modified t-test. The probability of each sample belonging to the control population was calculated, representing the percentage of samples from the control population that have lower value than a tested sample (Crawford \u0026amp; Garthwaite, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). The second experiment consisted in comparing CDI samples before and after FMT, as well as the stool donors\u0026rsquo; samples used for FMT, with control population of stool donor samples (n\u0026thinsp;=\u0026thinsp;113) from HCB.\u003c/p\u003e\u003cp\u003eFinally, the metabolic composition of stool donor\u0026rsquo;s samples from HCB (n\u0026thinsp;=\u0026thinsp;102) was analyzed to assess the correlation between the SCFAs and the abundance of bacterial taxa and alpha diversity indexes. For this, the Spearman\u0026rsquo;s method was applied between pairs of variables, and the \u003cem\u003ep-\u003c/em\u003evalues were adjusted for multiple testing using the Benjamini-Hochberg correction (\u003cem\u003eq\u003c/em\u003e-values).\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eData and code availability\u003c/h3\u003e\n\u003cp\u003eAll used datasets are publicly available at Qiita and NCBI BioProject repository, except from HCB sequences and metadata. Code produced for this analysis, alongside with raw and processed data files is available in a public repository (see data availability section below).\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCharacterization of AGP population\u0026rsquo;s alpha diversity\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe alpha diversity metrics obtained from the healthy samples of the AGP dataset (n = 1470) were used to describe the relationship between different metrics and to explore non-clinical factors affecting diversity. Table 1 contains the definitions and similarities between chosen alpha diversity metrics. In the literature these metrics are classified as estimations of richness, evenness or a combination of both. As expected, Pearson correlation proved that many of these indices are highly correlated (Figure 1a). Gini and Strong\u0026rsquo;s dominance indexes approach the estimation of alpha diversity in a reciprocal way compared to the other metrics (the higher the index, the lower the diversity/evenness). When inverted, both Gini and Strong\u0026rsquo;s index are closer to the group of indices that account for both richness and evenness (Shannon entropy and Simpsons\u0026rsquo;s index) or just evenness (Pielou\u0026rsquo;s evenness) component of diversity. By applying Exploratory Factor Analysis (psych package in R) we confirmed that there are two underlying latent factors (Supplementary Fig. S1), likely richness and evenness, that are driving this grouping of indices.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFurthermore, the distributions of alpha metrics were tested for normality using the Shapiro\u0026ndash;Wilk test. The skewness and kurtosis were also computed for all metrics (Supplementary Table S4). Although none of the metrics in our analysis satisfied the criteria for normality, richness-related metrics showed closer resemblance to a normal distribution (Figure 1b).\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"623\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 1.\u0026nbsp;\u003c/strong\u003eDefinitions of alpha diversity indices used in this work\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMetric\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDefinition\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eReferences\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRichness metrics\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eChao 1\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eEstimator of species richness, including rare or unobserved species\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Chao, 1984)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eMargalef\u0026rsquo;s index\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eRation of number of species in the community relative to the total number of individuals\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Magurran, 2004)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eMenhinick\u0026rsquo;s index\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eRatio of number of species to square root of number of individuals in the sample\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Magurran, 2004)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eFisher\u0026rsquo;s alpha\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eMeasuring the relationship between the number of species and the relative abundance of each species\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Fisher et al., 1943)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eFaith\u0026rsquo;s phylogenetic diversity\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eThe sum of OUT branch lengths. It takes into account phylogenetic distance between OUTs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Faith, 1992)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEvenness metrics\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eGini index\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eMeasure of evenness or inequality in the distribution of microbial species within a community. High Gini index \u0026ndash; high inequality in the distribution of species (due to one highly dominant species or a few dominant species with many rare species)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Gini, 1912)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eStrong\u0026rsquo;s dominance index\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eAbundance unevenness or dominance concentration. High SDI specifically indicates that the most abundant species is highly dominant, even if there are other species present in smaller quantities\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Strong, 2002)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003ePielou evenness\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eMeasure of relative evenness of species richness. Compares the observed diversity to the maximum possible diversity \u0026ndash; normalized version of Shannon\u0026rsquo;s index\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Pielou, 1966)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBoth richness and evenness\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eShannon\u0026rsquo;s entropy\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eHeterogeneity of a sample, redundancy, entropy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Shannon \u0026amp; Weaver, 1949)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eSimpson\u0026rsquo;s index\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 378px;\"\u003e\n \u003cp\u003eProbability that any two organisms sampled will be the sample phylotype\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e(Simpson, 1949)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003eMore detailed explanations of each metric with their mathematical formulas can be found in Supplementary section 1.2.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTo determine the non-clinical factors affecting alpha diversity within the healthy population, first we compared numerical features (age, height, weight, and BMI) with alpha diversity metrics and found no significant correlation. For categorical variables we attempted to estimate feature importance using a Random Forest classifier (Supplementary Fig. S2). The most important categories in both approaches were country of residence and country of birth, suggesting that geographically different populations might differ in alpha diversity. Further investigation confirmed that participants from European countries (both by birth and residence) had a higher average richness and evenness than those from North American countries. Additionally, participants of Asian origin (country of birth) had the lowest diversity and significantly differed from European samples in most metrics (Supplementary Fig. S3 and S4).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAlpha diversity comparison between control and case populations\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOne of the aims of this analysis was to assess whether and to what extent we can capture the differences in microbial communities between populations of healthy controls and samples of patients with diagnosed IBD or CDI, by only looking at their alpha diversity.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDifference in alpha diversity between IBD samples and control\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAlpha diversity values obtained from two studies related to IBD (Lloyd-Price et al., 2019; Webside: Qiita ID 11549) were compared to those calculated for the AGP control group. Two IBD conditions, Crohn\u0026rsquo;s disease - CD (n = 26) and ulcerative colitis - UC (n = 40) showed lower richness, but higher evenness when compared with the control samples. The exceptions are Faith\u0026rsquo;s PD (significant difference with healthy samples; p \u0026lt; 0.05) and Menhinick (not significant with p = 0.1) based on which CD had the highest richness (Supplementary Fig. S5). Furthermore, UC showed significant differences from the control population in more alpha metrics than CD (Supplementary Tables S5 and S6). This can be due to the heterogeneous severity of Crohn\u0026rsquo;s disease diagnosis or a low number of available samples. However, information about the severity was not provided in metadata.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo increase the sample size and gain further insights, data from a longitudinal CD study (V\u0026aacute;zquez-Baeza et al., 2018) were added to the analysis (n = 293). Control samples from healthy family members were included in this dataset (n = 353). This control population had a significantly different distribution mean from the AGP control population for most of the alpha metrics (Supplementary Table S7), as did the CD samples from previous IBD datasets (Supplementary Table S8). Therefore, we proceeded with the analysis with the original study groups. We confirmed the following conclusions from the original study: a) Crohn\u0026rsquo;s disease samples have significantly lower richness and evenness compared to controls (Supplementary Table S9); b) CD samples that previously underwent some surgical intervention had lower alpha diversity than those that did not (Supplementary Table S10, Supplementary Fig. S6). This difference was significant based on the Mann-Whitney-Wilcoxon test for more alpha metrics than those in the first CD dataset.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDifference in alpha diversity between CDI samples and control and effect of FMT on alpha diversity\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAlpha diversity of CDI samples (n = 73) from Khanna et al. (2016) showed highly significant differences from AGP healthy samples, showing lower richness and lower evenness (Supplementary Table S11, Supplementary Fig. S7). This condition is the only one treated with FMT. Studies examining the success of this type of treatment usually estimate it by comparing the composition of the microbial community before and after treatment. However, we wanted to quantify the improvement of the microbial community after FMT through alpha diversity. For this we examined a longitudinal study of 4 CDI patients treated with FMT whose progress was examined in multiple time points after procedure yielding 88 post-FMT samples (Weingarden et al., 2015). We noticed a consistent trend of increased microbial richness and evenness over time after the procedure in all subjects (Figure 2). Although the post-FMT diversity values varied over time, all time points after FMT showed higher richness and lower dominance than the initial value before treatment.\u003c/p\u003e\n\u003cp\u003eFurthermore, we examined whether the same trend of worse response of CDI patients to FMT in cases with an underlying IBD suggested by Khanna et al. (2017) could be observed on the alpha diversity level. Supplementary Fig. S9 shows the trend of improvement in alpha diversity on days 7th and 28th day after FMT. For all alpha indices, the diversity increases toward the donor\u0026apos;s mean value. However, samples with underlying CD showed a richness and evenness decrease on the 28th day compared to 7th day for all alpha metrics (except for Pielou index).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDifference in alpha diversity between CDI and healthy donor samples from Hospital Cl\u0026iacute;nic de Barcelona\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe compared alpha diversity in CDI samples before (n = 18) and after transplantation (n = 38), as well as the difference between CDI patients and healthy stool donors (n = 151). The results show a significant difference between CDI samples before FMT and stool donor samples for all alpha metrics (Figure 3). Furthermore, FMT increased samples\u0026rsquo; richness (significantly for all richness metrics except form Faith PD) and evenness (lower Gini and Strong, higher Shannon, Pielou, and Simpson although for the last two the difference was not significant; Supplementary Table S12). Additionally, we also noticed that donor samples from the HCB have higher richness (except from Menhinick) and much higher evenness than AGP samples that we have previously used as controls (Figure 4).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eStatistical power analysis\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKers \u0026amp; Saccenti (2022) highlighted the difference in the statistical power of various alpha metrics when comparing the two groups. Wilcoxon effect size was used to determine which metric from each of the groups (richness and evenness) is the most sensitive to the difference between healthy and unhealthy samples.\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"623\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 2.\u0026nbsp;\u003c/strong\u003eDefinitions of alpha diversity indices used in this work\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eParameter\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003ep\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e-value adjusted\u003csup\u003ea\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEffect size (r)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eConfidence low\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eConfidence high\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMagnitude\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eGini index (x)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026lt;10\u003csup\u003e-16\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.553\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eLarge\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eChao1 (+)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026lt;10\u003csup\u003e-16\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.400\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eModerate\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMenhinick\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026lt;10\u003csup\u003e-16\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.377\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eModerate\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMargalef (+)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026lt;10\u003csup\u003e-16\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.375\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eModerate\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eFisher alpha (+)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026lt;10\u003csup\u003e-16\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.375\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eModerate\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eFaith PD (+)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026lt;10\u003csup\u003e-16\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.302\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eModerate\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eShannon entropy (*)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026lt;10\u003csup\u003e-16\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.249\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eSmall \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eSimpson (*)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026lt;10\u003csup\u003e-16\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.189\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eSmall\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003ePielou evenness (x)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1,19e\u003csup\u003e-10\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.135\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eSmall\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eStrong (x)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.0022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.064\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eSmall\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003csup\u003ea\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e-value adjusted is a result of the Mann-Whitney-Wilcoxon test for the difference between means of all healthy samples (n = 1823) and unhealthy sample (n = 432). Adjustment is done using the false discovery rate (FDR) method. Wilcoxon effect size (r) with estimated confidence interval varies from 0 to 1 and is commonly interpreted as: 0.10 \u0026ndash; 0.3 (small effect), 0.30 \u0026ndash; 0.5 (moderate effect) and \u0026gt;= 0.5 (large effect). (+) richness metric, (x) evenness metric, (*) both richness and evenness.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eBased on the results shown in Table 2, the Gini index seems to show the highest effect size when comparing healthy and unhealthy samples, scoring better than the other metrics in the \u0026ldquo;evenness\u0026rdquo; group. Chao1 is the second best, and the best among the \u0026ldquo;richness\u0026rdquo; metrics.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eFeature importance and accuracy of Random Forest classifier\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe compared the classifier\u0026apos;s performance on different datasets to determine if it was more sensitive to one disease than another. Additionally, we investigated whether the classifier was better at predicting the overall health status (healthy or unhealthy) of the samples or predicting specific conditions. The highest accuracy was obtained with a model containing all alpha metrics when it was trained on all datasets together (~88% for predicting condition, ~89% for health status; Supplementary Table S13), and in the case of IBD and the healthy dataset alone (~87% for predicting condition, ~88% for health status; Supplementary Table S14). However, the decrease in accuracy is not substantial when we choose only two metrics (one representing the richness metric and the other the evenness metric). The highest accuracy was achieved when the model used the Gini index as an evenness metric (82-85% for predicting condition, 84\u0026ndash;86% for health status).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEven higher accuracy was obtained when the random forest classifier was trained on the healthy and CDI datasets using different models (Supplementary Table S15). This classifier appears to be able to determine the differences between healthy and CDI samples with more than 90% accuracy in all the models, with each model containing Gini index reaching 100% accuracy (except for the Gini-Faith PD model with 99.8% accuracy). Similar results were obtained from the classification of HCB CDI and healthy donor samples (Supplementary Table S16). In this case, more models showed absolute accuracy, but we also had fewer samples for classification in this dataset. Again, models including the Gini index scored the best, with the addition of two models: Menhinick-Strong and Menhinick-Shannon, that exhibited the same accuracy. For all classifiers, models with the Gini index performed the best. Therefore, the Gini index scored the highest importance for Random Forest classifier (Supplementary Fig. S10). The best \u0026ldquo;richness\u0026rdquo; metric varied between different classifiers, although the accuracy was not affected much when the \u0026ldquo;evenness\u0026rdquo; metric was Gini index.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eModified t-test for single case comparisons\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe probability of each test population sample belonging to the control (AGP) population from the first experiment is summarized in Table 3. On average, healthy samples exhibited a higher probability of belonging to the control population compared to CDI samples. \u0026ldquo;Richness\u0026rdquo; metrics showed great results with all CDI samples located in the 5% (Faith PD) to 12% (Fisher alpha) of the control population. In contrast, \u0026ldquo;evenness\u0026rdquo; metrics showed significant overlap between CDI and healthy test samples, with the exception of the Gini index, which achieved complete separation. \u0026nbsp; These findings indicate that \u0026quot;evenness\u0026quot; metrics, when analyzed using t-tests, may not provide sufficient discriminatory capability to distinguish between healthy and unhealthy samples.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"623\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 3.\u0026nbsp;\u003c/strong\u003e\u003cem\u003ep\u003c/em\u003e-values obtained by modified t-test: AGP as control population\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" style=\"width: 156px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDiversity index\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" style=\"width: 312px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eProbability [Mean (Min, Max)]\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 156px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eN healthy overlap CDI*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 156px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCDI (n = 73)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 156px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHealthy (n = 437)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003eChao1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.026 (0.005, 0.082)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.477 (0.013, 1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e27 (6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003eMargalef\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.030 (0.004, 0.116)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.481 (0.008, 1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e54 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003eMenhinick\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.025 (0.004, 0.092)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.481 (0.008, 1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e40 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003eFisher alpha\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.043 (0.014, 0.122)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.474 (0.020, 1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e53 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003eFaith PD\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.008 (0.001, 0.024)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.482 (0.005, 0.999)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e5 (1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003eGini index\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0 (0, 0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.531 (0.0002, 0.948)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0 (0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003eStrong\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.480 (0.016, 0.973)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.494 (0.0024, 0.999)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e410 (94%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003ePielou evenness\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.405 (0.010, 0.887)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.520 (0.0001, 0.915)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e429 (98%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003eShannon entropy\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.180 (0.003, 0.628)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.511 (0.0004, 0.965)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e246 (56%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003eSimpson\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.371 (0.001, 0.728)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e0.536 (0, 0.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 156px;\"\u003e\n \u003cp\u003e315 (72%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e-value mean, maximum and minimum were calculated from the results of applying modified t-test on \u0026ldquo;test\u0026rdquo; population. T value was calculated based on Crawford \u0026amp; Howell (1998) formula. The obtained value was compared to t-distribution with one tailed t-test observing lower end. *How many healthy samples have lower probability of belonging to the control population than the CDI sample with the highest probability.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eBased on Table 4, there is an increase in the probability of FMT recipients and donor samples belonging to the HCB donor population compared to CDI samples before FMT. The \u0026ldquo;richness\u0026rdquo; metrics, along with the Gini index, demonstrated better performance in distinguishing between healthy and unhealthy samples. Among these metrics, the Chao1 index exhibited the least overlap between donor samples and CDI before FMT, while showing the greatest overlap with CDI after FMT- an outcome aligned with expectations. The Gini index also provided good discriminatory power, achieving complete separation for donor samples and CDI samples before FMT.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"623\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"top\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 4.\u0026nbsp;\u003c/strong\u003e\u003cem\u003ep-\u003c/em\u003evalues obtained by modified t-test: stool donors from HCB as control population\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDiversity index\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" style=\"width: 312px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eProbability [Mean (Min, Max)]\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCDI pre overlap\u0026nbsp;\u003c/strong\u003e\u0026Dagger;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCDI post overlap\u0026nbsp;\u003c/strong\u003e\u0026Dagger;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCDI pre (n = 18)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCDI post (n = 38)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDonors (n = 38)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003eChao1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.032\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(0, 0.218)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.206\u003c/p\u003e\n \u003cp\u003e(0, 0.995)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.562\u003c/p\u003e\n \u003cp\u003e(0.018, 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2 (5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e26 (68%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003eMargalef\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.044\u003c/p\u003e\n \u003cp\u003e(0, 0.328)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.220\u003c/p\u003e\n \u003cp\u003e(0, 0.983)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.574\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(0.016, 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2 (5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e12 (32%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003eMenhinic k\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.242\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(0, 0.948)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.523\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(0.001, 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.616\u003c/p\u003e\n \u003cp\u003e(0.016, 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2 (5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e12 (32%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003eFisher alpha\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.062\u003c/p\u003e\n \u003cp\u003e(0, 0.430)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.263\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(0.001, 0.998)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.586\u003c/p\u003e\n \u003cp\u003e(0.021, 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2 (5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e21 (55%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003eGini index*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003cp\u003e(0, 0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.0003\u003c/p\u003e\n \u003cp\u003e(0, 0.003)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.027\u003c/p\u003e\n \u003cp\u003e(0, 0.488)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0 (0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e19 (50%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003eStrong*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.091\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(0, 0.860)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.090\u003c/p\u003e\n \u003cp\u003e(0, 0.888)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.181\u003c/p\u003e\n \u003cp\u003e(0, 0.679)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e38 (100%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e38 (100%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003ePielou evenness\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.078\u003c/p\u003e\n \u003cp\u003e(0, 0.650)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.083\u003c/p\u003e\n \u003cp\u003e(0, 0.628)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.486\u003c/p\u003e\n \u003cp\u003e(0.003, 0.949)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e32 (84%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e35 (92%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003eShannon entropy\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.011 (0, 0.095)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.089\u003c/p\u003e\n \u003cp\u003e(0, 0.798)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.535\u003c/p\u003e\n \u003cp\u003e(0.001, 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2 (5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e20 (53%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003eSimpson\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.041\u003c/p\u003e\n \u003cp\u003e(0, 0.395)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.103\u003c/p\u003e\n \u003cp\u003e(0, 0.822)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.569\u003c/p\u003e\n \u003cp\u003e(0, 0.983)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e4 (10%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e21 (55%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e*For Gini index and Strong dominance probability is calculated for the upper tail of T distribution since healthy samples have lower values. \u0026Dagger; How many donor samples have lower probability of belonging the control population than the CDI sample before and after FMT with the highest probability. Desirable outcomes would be lower overlap with CDI pre and higher overlap with CDI post.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eSCFAs correlation with bacterial genera and alpha-diversity indexes\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eResults of the metabolic composition (SCFAs) of stool donors\u0026rsquo; samples from HCB were correlated with the abundance of bacterial taxa and alpha diversity indexes. Of the 78 bacterial genera detected in the metagenomic study, 8 showed correlation with one of the main SCFAs (\u003cem\u003eq\u003c/em\u003e-value \u0026lt; 0.01). Higher SCFAs levels in stool were associated with bacterial genera responsible for producing SCFAs, such as \u003cem\u003eAnaerostipes\u003c/em\u003e, \u003cem\u003eBacteroides\u003c/em\u003e, \u003cem\u003eCoprococcus\u003c/em\u003e, \u003cem\u003eFaecalibacterium\u003c/em\u003e, \u003cem\u003eOdoribacter\u003c/em\u003e, and \u003cem\u003eRoseburia\u003c/em\u003e. The AF12 genera, not known for its capacity to produce SCFAs, also presented a positive correlation with acetate, although very weak. On the other hand, \u003cem\u003eOxalobacter\u003c/em\u003e inversely correlated with fecal butyrate (Figure 5a).\u003c/p\u003e\n\u003cp\u003eRegarding alpha diversity indexes, SCFAs showed a significant inverse correlation with 8 out of the 10 alpha diversity indexes assessed (\u003cem\u003eq\u003c/em\u003e-value \u0026lt; 0.01), being fecal butyrate levels the ones that present more significant correlations. Only Gini\u0026rsquo;s index (which measures the inequality in the distribution of microbial species) shows a positive correlation with butyrate. Strong\u0026rsquo;s dominance index showed no significant correlation with any of SCFAs analyzed (Figure 5b).\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eGut microbiome diversity is increasingly recognized as a key indicator of intestinal health. However, a consistent definition of dysbiosis remains elusive due to its conceptual ambiguity. Benchmarking alpha diversity metrics against well-characterized healthy and diseased cohorts offers a valuable strategy for assessing their clinical relevance. Such benchmarks provide evidence that these metrics can reliably reflect microbial disturbances. In particular, richness and evenness indices show strong alignment with gut health, supporting their utility in both clinical and research contexts. Nonetheless, this study underscores important limitations in interpreting alpha diversity metrics, especially when short-chain fatty acids (SCFAs) are used as indirect proxies for microbial diversity.\u003c/p\u003e\u003cp\u003eOur findings demonstrate that healthy gut microbial communities are consistently characterized by higher richness and evenness compared to unhealthy states such as inflammatory bowel disease (IBD) or \u003cem\u003eC. difficile\u003c/em\u003e infection (CDI). Notably, CDI samples displayed robust and significant reductions in richness, with recovery observed post-fecal microbiota transplant (FMT), reinforcing the role of alpha diversity metrics in tracking intervention efficacy. However, the higher richness observed in the Lloyd-Price et al. (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) CDI samples, compared to AGP controls highlights the challenges posed by disease heterogeneity and the importance of using well-characterized control groups. c\u003c/p\u003e\u003cp\u003eWhile alpha diversity is a powerful summary measure, it is important to acknowledge its limitations. Metrics such as Chao1 and Faith's PD quantify richness but fail to capture microbial composition and functional interactions. By contrast, the Gini index, an underutilized evenness metric, emerged as a robust discriminator between healthy and unhealthy samples across datasets. Combining richness and evenness metrics, as demonstrated by our predictive models, significantly improved accuracy, particularly in distinguishing severe dysbiosis such as CDI.\u003c/p\u003e\u003cp\u003eSCFAs, including butyrate and acetate, are often viewed as beneficial markers of microbial functionality. However, our results highlight a counterintuitive inverse correlation between SCFA levels and alpha diversity metrics. This aligns with prior studies (De La Cuesta-Zuluaga et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Jones et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), suggesting that SCFA abundance reflects the activity of specific taxa (e.g., \u003cem\u003eFaecalibacterium, Anaerostipes, Roseburia\u003c/em\u003e), rather than the overall diversity of the ecosystem. Furthermore, factors such as SCFA absorption in the colon (90\u0026ndash;95%), daily food choices, transit time, and the time of stool collection introduce variability, undermining stool SCFA concentrations as a direct indicator of microbial production or diversity (Johnson et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Jones et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; McNeil et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1978\u003c/span\u003e; Vogt \u0026amp; Wolever, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). These findings emphasize that while SCFAs play a critical role in gut health, they cannot serve as standalone markers for microbial diversity. Alpha diversity metrics, which consider all taxa, provide a broader snapshot of ecosystem structure, albeit without functional specificity.\u003c/p\u003e\u003cp\u003eSeveral limitations warrant consideration. The absence of high-quality, publicly available control datasets with comprehensive clinical metadata limits the generalizability of our findings. For instance, the AGP dataset, while extensive, relies on self-reported health statuses, which may not accurately reflect microbiome profiles. Conversely, well-characterized donor populations, such as those from HCB, are limited in size but hold promise for large-scale validation studies. Geographical variability introduces additional confounding factors, with differences in lifestyle and diet influencing diversity metrics. This underscores the need for population-specific controls when interpreting alpha diversity benchmarks.\u003c/p\u003e\u003cp\u003eDespite the challenges in defining dysbiosis, our study demonstrates that alpha diversity metrics, particularly combinations of richness and evenness indices, are valuable markers of gut microbiota health. Metrics like the Gini index, Chao1, and Faith\u0026rsquo;s PD provide robust, complementary insights into microbial diversity, while SCFA abundance, although functionally significant and informative, remains an imperfect proxy of microbial diversity. Addressing current limitations, including dataset quality and population variability, will be essential for advancing the clinical application of these metrics in diagnosing and monitoring gut microbiome disturbances.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eEthics approval and consent to participate\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe analysis of samples from HCB patients was approved by the Ethics Committee of Research with medicines (CEIm) of HCB (Ref HCB/2016/0824 and Ref HCB/2024/0239), with appropriate informed consent from all participants in compliance with the Declaration of Helsinki and applicable national and European legislation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eConsent for publication\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eFunding\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study has recived support from the grant CEX2023-0001290-S funded by MCIN/AEI/ 10.13039/501100011033, and support from the Generalitat de Catalunya through the CERCA Program. ASA was supported by the PREDOCS-UB 2022 grant from the University of Barcelona (UB).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAvailability of data and materials\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData for the analyses in this study included datasets from public repositories (see the Methods section) and data from patients at HCB. Raw sequencing reads from HCB patients are available in NCBI Sequence Read Archive (SRA) under BioProject accession PRJNA1337461, with Biosample accessions SAMN52298702-SAMN52298908, and SRA run accessions SRR35728543-SRR35728749. All data are publicly accessible at: https://www.ncbi.nlm.nih.gov/sra/PRJNA1337461\u003c/p\u003e\n\u003cp\u003eAll the alpha-diversity data and code required to reproduce the analysis presented in this paper, including links to the original repositories, is available in the following GitHub repository: https://github.com/CDB-coreBM/mortvanski_et_al_2025.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAcknowledgments\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe acknowledge the Departments of Microbiology and Biochemistry and Molecular Genetics for their contribution.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAira, A., Rubio, E., Ruiz, A., Vergara, A., Casals-Pascual, C., Rico, V., Su\u0026ntilde;\u0026eacute;-Negre, J.M., \u0026amp; Soriano, A. (2022). New Procedure to Maintain Fecal Microbiota in a Dry Matrix Ready to Encapsulate. \u003cem\u003eFront. Cell. Infect. Microbiol.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 899257.\u003c/li\u003e\n\u003cli\u003eAlva-Murillo, N., Ochoa-Zarzosa, A., \u0026amp; L\u0026oacute;pez-Meza, J.E. (2012). Short chain fatty acids (propionic and hexanoic) decrease Staphylococcus aureus internalization into bovine mammary epithelial cells and modulate antimicrobial peptide expression. \u003cem\u003eVet. Microbiol.\u003c/em\u003e \u003cstrong\u003e155\u003c/strong\u003e, 324\u0026ndash;331.\u003c/li\u003e\n\u003cli\u003eAmir, A., McDonald, D., Navas-Molina, J.A., Kopylova, E., Morton, J.T., Zech Xu, Z., Kightley, E.P., Thompson, L.R., Hyde, E.R., Gonzalez, A., \u0026amp; Knight, R. (2017). Deblur Rapidly Resolves Single-Nucleotide Community Sequence Patterns. \u003cem\u003emSystems\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, e00191-16.\u003c/li\u003e\n\u003cli\u003eBelkaid, Y., \u0026amp; Hand, T.W. (2014). Role of the Microbiota in Immunity and Inflammation. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e157\u003c/strong\u003e, 121\u0026ndash;141.\u003c/li\u003e\n\u003cli\u003eBokulich, N.A., Kaehler, B.D., Rideout, J.R., Dillon, M., Bolyen, E., Knight, R., Huttley, G.A., \u0026amp; Gregory Caporaso, J. (2018). Optimizing taxonomic classification of marker-gene amplicon sequences with QIIME 2\u0026rsquo;s q2-feature-classifier plugin. \u003cem\u003eMicrobiome\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 90.\u003c/li\u003e\n\u003cli\u003eBolyen, E., Rideout, J.R., Dillon, M.R., et al. (2019). Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 852\u0026ndash;857.\u003c/li\u003e\n\u003cli\u003eCallahan, B.J., McMurdie, P.J., Rosen, M.J., Han, A.W., Johnson, A.J.A., \u0026amp; Holmes, S.P. (2016). DADA2: High resolution sample inference from Illumina amplicon data. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 581\u0026ndash;583.\u003c/li\u003e\n\u003cli\u003eCarding, S., Verbeke, K., Vipond, D.T., Corfe, B.M., \u0026amp; Owen, L.J. (2015). Dysbiosis of the gut microbiota in disease. \u003cem\u003eMicrob. Ecol. Health Dis.\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eChao, A. (1984). Nonparametric Estimation of the Number of Classes in a Population. \u003cem\u003eScand. J. Stat.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 265\u0026ndash;270.\u003c/li\u003e\n\u003cli\u003eCrawford, J.R., \u0026amp; Garthwaite, P.H. (2005). Testing for Suspected Impairments and Dissociations in Single-Case Studies in Neuropsychology: Evaluation of Alternatives Using Monte Carlo Simulations and Revised Tests for Dissociations. \u003cem\u003eNeuropsychology\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 318\u0026ndash;331.\u003c/li\u003e\n\u003cli\u003eCrawford, J.R., \u0026amp; Howell, D.C. (1998). Comparing an Individual\u0026rsquo;s Test Score Against Norms Derived from Small Samples. \u003cem\u003eClin. Neuropsychol.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 482\u0026ndash;486.\u003c/li\u003e\n\u003cli\u003eCulp, E.J., \u0026amp; Goodman, A.L. (2023). Cross-feeding in the gut microbiome: Ecology and mechanisms. \u003cem\u003eCell Host Microbe\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 485\u0026ndash;499.\u003c/li\u003e\n\u003cli\u003eDe La Cuesta-Zuluaga, J., Mueller, N., \u0026Aacute;lvarez-Quintero, R., Vel\u0026aacute;squez-Mej\u0026iacute;a, E., Sierra, J., Corrales-Agudelo, V., Carmona, J., Abad, J., \u0026amp; Escobar, J. (2018). Higher Fecal Short-Chain Fatty Acid Levels Are Associated with Gut Microbiome Dysbiosis, Obesity, Hypertension and Cardiometabolic Disease Risk Factors. \u003cem\u003eNutrients\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 51.\u003c/li\u003e\n\u003cli\u003eFaith, D.P. (1992). Conservation evaluation and phylogenetic diversity. \u003cem\u003eBiol. Conserv.\u003c/em\u003e \u003cstrong\u003e61\u003c/strong\u003e, 1\u0026ndash;10.\u003c/li\u003e\n\u003cli\u003eFisher, R.A., Corbet, A.S., \u0026amp; Williams, C.B. (1943). The Relation Between the Number of Species and the Number of Individuals in a Random Sample of an Animal Population. \u003cem\u003eJ. Anim. Ecol.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 42.\u003c/li\u003e\n\u003cli\u003eGill, S.K., Rossi, M., Bajka, B., \u0026amp; Whelan, K. (2021). Dietary fibre in gastrointestinal health and disease. \u003cem\u003eNat. Rev. Gastroenterol. Hepatol.\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 101\u0026ndash;116.\u003c/li\u003e\n\u003cli\u003eGini, C. (1912). Variabilita e Mutabilita contributo allo studio delle distribuzioni e delle relazioni statistiche. (Bologna: Tipografia di Paolo Cuppini).\u003c/li\u003e\n\u003cli\u003eGonzalez, A., Navas-Molina, J.A., Kosciolek, T., et al. (2018). Qiita: rapid, web-enabled microbiome meta-analysis. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 796\u0026ndash;798.\u003c/li\u003e\n\u003cli\u003eHagerty, S.L., Hutchison, K.E., Lowry, C.A., \u0026amp; Bryan, A.D. (2020). An empirically derived method for measuring human gut microbiome alpha diversity: Demonstrated utility in predicting health-related outcomes among a human clinical sample. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, e0229204.\u003c/li\u003e\n\u003cli\u003eHMP Consortium (2012). Structure, Function and Diversity of the Healthy Human Microbiome. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e486\u003c/strong\u003e, 207\u0026ndash;214.\u003c/li\u003e\n\u003cli\u003eJohnson, K.V.-A., \u0026amp; Burnet, P.W.J. (2016). Microbiome: Should we diversify from diversity? \u003cem\u003eGut Microbes\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 455\u0026ndash;458.\u003c/li\u003e\n\u003cli\u003eJohnson, A.J., Vangay, P., Al-Ghalith, G.A., Hillmann, B.M., Ward, T.L., Shields-Cutler, R.R., Kim, A.D., Shmagel, A.K., Syed, A.N., Walter, J., Menon, R., Koecher, K., \u0026amp; Knights, D. (2019). Daily Sampling Reveals Personalized Diet-Microbiome Associations in Humans. \u003cem\u003eCell Host Microbe\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 789-802.e5.\u003c/li\u003e\n\u003cli\u003eJones, J., Reinke, S.N., Ali, A., Palmer, D.J., \u0026amp; Christophersen, C.T. (2021). Fecal sample collection methods and time of day impact microbiome composition and short chain fatty acid concentrations. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 13964.\u003c/li\u003e\n\u003cli\u003eKatoh, K. (2002). MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 3059\u0026ndash;3066.\u003c/li\u003e\n\u003cli\u003eKers, J.G., \u0026amp; Saccenti, E. (2022). The Power of Microbiome Studies: Some Considerations on Which Alpha and Beta Metrics to Use and How to Report Results. \u003cem\u003eFront. Microbiol.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 796025.\u003c/li\u003e\n\u003cli\u003eKhanna, S., Montassier, E., Schmidt, B., Patel, R., Knights, D., Pardi, D.S., \u0026amp; Kashyap, P.C. (2016). Gut microbiome predictors of treatment response and recurrence in primary \u003cem\u003eClostridium difficile\u003c/em\u003e infection. \u003cem\u003eAliment. Pharmacol. Ther.\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e, 715\u0026ndash;727.\u003c/li\u003e\n\u003cli\u003eKhanna, S., Vazquez-Baeza, Y., Gonz\u0026aacute;lez, A., et al. (2017). Changes in microbial ecology after fecal microbiota transplantation for recurrent C. difficile infection affected by underlying inflammatory bowel disease. \u003cem\u003eMicrobiome\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 55.\u003c/li\u003e\n\u003cli\u003eKim, C.H. (2023). Complex regulatory effects of gut microbial short-chain fatty acids on immune tolerance and autoimmunity. \u003cem\u003eCell. Mol. Immunol.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 341\u0026ndash;350.\u003c/li\u003e\n\u003cli\u003eKoh, A., De Vadder, F., Kovatcheva-Datchary, P., \u0026amp; B\u0026auml;ckhed, F. (2016). From Dietary Fiber to Host Physiology: Short-Chain Fatty Acids as Key Bacterial Metabolites. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e165\u003c/strong\u003e, 1332\u0026ndash;1345.\u003c/li\u003e\n\u003cli\u003eLathakumari, R.H., Vajravelu, L.K., Satheesan, A., Ravi, S., \u0026amp; Thulukanam, J. (2024). Antibiotics and the gut microbiome: Understanding the impact on human health. \u003cem\u003eMed. Microecol.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 100106.\u003c/li\u003e\n\u003cli\u003eLi, Z., Zhou, J., Liang, H., Ye, L., Lan, L., Lu, F., Wang, Q., Lei, T., Yang, X., Cui, P., \u0026amp; Huang, J. (2022). Differences in Alpha Diversity of Gut Microbiota in Neurological Diseases. \u003cem\u003eFront. Neurosci.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 879318.\u003c/li\u003e\n\u003cli\u003eLloyd-Price, J., Arze, C., Ananthakrishnan, A.N., et al. (2019). Multi-omics of the gut microbial ecosystem in inflammatory bowel diseases. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e569\u003c/strong\u003e, 655\u0026ndash;662.\u003c/li\u003e\n\u003cli\u003eLupp, C., Robertson, M.L., Wickham, M.E., Sekirov, I., Champion, O.L., Gaynor, E.C., \u0026amp; Finlay, B.B. (2007). Host-Mediated Inflammation Disrupts the Intestinal Microbiota and Promotes the Overgrowth of Enterobacteriaceae. \u003cem\u003eCell Host Microbe\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 119\u0026ndash;129.\u003c/li\u003e\n\u003cli\u003eMagurran, A.E. (2004). Measuring Biological Diversity. \u003cem\u003eBlackwell\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eManor, O., Dai, C.L., Kornilov, S.A., Smith, B., Price, N.D., Lovejoy, J.C., Gibbons, S.M., \u0026amp; Magis, A.T. (2020). Health and disease markers correlate with gut microbiome composition across thousands of people. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 5206.\u003c/li\u003e\n\u003cli\u003eMcDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A., Andersen, G.L., Knight, R., \u0026amp; Hugenholtz, P. (2012). An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. \u003cem\u003eISME J.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 610\u0026ndash;618.\u003c/li\u003e\n\u003cli\u003eMcDonald, D., Hyde, E., Debelius, J.W., et al. (2018). American Gut: an Open Platform for Citizen Science Microbiome Research. \u003cstrong\u003e3\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eMcNeil, N.I., Cummings, J.H., \u0026amp; James, W.P. (1978). Short chain fatty acid absorption by the human large intestine. \u003cem\u003eGut\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 819\u0026ndash;822.\u003c/li\u003e\n\u003cli\u003eMosca, A., Leclerc, M., \u0026amp; Hugot, J.P. (2016). Gut Microbiota Diversity and Human Diseases: Should We Reintroduce Key Predators in Our Ecosystem? \u003cem\u003eFront. Microbiol.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003ePielou, E.C. (1966). The Measurement of Diversity in Different Types of Biological Colledions. \u003cem\u003eJ. Theor. Biol.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 131\u0026ndash;144.\u003c/li\u003e\n\u003cli\u003ePlassais, J., Gbikpi-Benissan, G., Figarol, M., Scheperjans, F., Gorochov, G., Derkinderen, P., \u0026amp; Cervino, A.C.L. (2021). Gut microbiome alpha-diversity is not a marker of Parkinson\u0026rsquo;s disease and multiple sclerosis. \u003cem\u003eBrain Commun.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, fcab113.\u003c/li\u003e\n\u003cli\u003ePrice, M.N., Dehal, P.S., \u0026amp; Arkin, A.P. (2009). FastTree: Computing Large Minimum Evolution Trees with Profiles instead of a Distance Matrix. \u003cem\u003eMol. Biol. Evol.\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 1641\u0026ndash;1650.\u003c/li\u003e\n\u003cli\u003eSayol-Altarriba, A., Aira, A., Villasante, A., Albarrac\u0026iacute;n, R., Faneca, J., Casals, G., Villanueva-Ca\u0026ntilde;as, J.L., \u0026amp; Casals-Pascual, C. (2024). Normalization of short-chain fatty acid concentration by bacterial count of stool samples improves discrimination between eubiotic and dysbiotic gut microbiota caused by \u003cem\u003eClostridioides difficile\u003c/em\u003e infection-. \u003cem\u003eGut Microbes\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 2415488.\u003c/li\u003e\n\u003cli\u003eShannon, C.E., \u0026amp; Weaver, W. (1949). The Mathematical Theory of Communication (Urbana, IL: The University of Illinois Press).\u003c/li\u003e\n\u003cli\u003eSimpson, E. (1949). Measurement of Diversity. \u003cstrong\u003e163\u003c/strong\u003e, 688.\u003c/li\u003e\n\u003cli\u003eStrong, W.L. (2002). Assessing species abundance unevenness within and between plant communities. \u003cem\u003eCommunity Ecol.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 237\u0026ndash;246.\u003c/li\u003e\n\u003cli\u003eSuzuki, T., Yoshida, S., \u0026amp; Hara, H. (2008). Physiological concentrations of short-chain fatty acids immediately suppress colonic epithelial permeability. \u003cem\u003eBr. J. Nutr.\u003c/em\u003e \u003cstrong\u003e100\u003c/strong\u003e, 297\u0026ndash;305.\u003c/li\u003e\n\u003cli\u003eThompson, L.R., Sanders, J.G., McDonald, D., et al. (2017). A communal catalogue reveals Earth\u0026rsquo;s multiscale microbial diversity. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e551\u003c/strong\u003e, 457\u0026ndash;463.\u003c/li\u003e\n\u003cli\u003eTurnbaugh, P.J., Hamady, M., Yatsunenko, T., et al. (2009). A core gut microbiome in obese and lean twins. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e457\u003c/strong\u003e, 480\u0026ndash;484.\u003c/li\u003e\n\u003cli\u003eV\u0026aacute;zquez-Baeza, Y., Gonzalez, A., Xu, Z.Z., Washburne, A., Herfarth, H.H., Sartor, R.B., \u0026amp; Knight, R. (2018). Guiding longitudinal sampling in IBD cohorts. \u003cem\u003eGut\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 1743\u0026ndash;1745.\u003c/li\u003e\n\u003cli\u003eVogt, J.A., \u0026amp; Wolever, T.M.S. (2003). Fecal Acetate Is Inversely Related to Acetate Absorption from the Human Rectum and Distal Colon. \u003cem\u003eJ. Nutr.\u003c/em\u003e \u003cstrong\u003e133\u003c/strong\u003e, 3145\u0026ndash;3148.\u003c/li\u003e\n\u003cli\u003eWebside: Qiita ID 11549 Qiita ID 11549. Metaomics Reveals Microbiome Based Proteolysis as a Driver of Ulcerative Colitis Severity.\u003c/li\u003e\n\u003cli\u003eWeingarden, A., Gonz\u0026aacute;lez, A., V\u0026aacute;zquez-Baeza, Y., Weiss, S., Humphry, G., Berg-Lyons, D., Knights, D., Unno, T., Bobr, A., Kang, J., Khoruts, A., Knight, R., \u0026amp; Sadowsky, M.J. (2015). Dynamic changes in short- and long-term bacterial composition following fecal microbiota transplantation for recurrent Clostridium difficile infection. \u003cem\u003eMicrobiome\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 10.\u003c/li\u003e\n\u003cli\u003eYao, L., Davidson, E.A., Shaikh, M.W., Forsyth, C.B., Prenni, J.E., \u0026amp; Broeckling, C.D. (2022). Quantitative analysis of short-chain fatty acids in human plasma and serum by GC\u0026ndash;MS. \u003cem\u003eAnal. Bioanal. Chem.\u003c/em\u003e \u003cstrong\u003e414\u003c/strong\u003e, 4391\u0026ndash;4399.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Gut microbiota, dysbiosis, microbial alpha diversity, Clostridioides difficile infection (CDI), inflammatory bowel disease (IBD), fecal microbiota transplantation (FMT), short-chain fatty acids (SCFAs)","lastPublishedDoi":"10.21203/rs.3.rs-7987343/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7987343/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cb\u003eBackground\u003c/b\u003e\u003c/p\u003e\u003cp\u003eSelecting appropriate alpha diversity metrics is essential for capturing the biological and clinical relevance of gut microbiome studies. However, no consensus or clear rationale currently guides this choice. In this study, we compared the distribution of 10 commonly used alpha diversity metrics in patients with inflammatory bowel disease (IBD) and \u003cem\u003eClostridioides difficile\u003c/em\u003e infection (CDI) to those in a healthy reference population. The healthy reference group consisted of fecal donor samples validated for clinical use. Our aim was to benchmark the performance of these metrics against a healthy gut microbiome.\u003c/p\u003e\u003cp\u003e\u003cb\u003eResults\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe found that improvements in health status were associated with increases in both richness and evenness components of alpha diversity. The ability to differentiate between health conditions varied among metrics. Notably, the Gini index emerged as a robust evenness metric for predicting health status and detecting group differences, while most richness metrics showed consistent trends across all comparisons.\u003c/p\u003e\u003cp\u003e\u003cb\u003eConclusions\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThese results suggest that alpha diversity metrics can serve as valuable tools to capture microbiome disturbances and monitoring gut microbiome health.\u003c/p\u003e","manuscriptTitle":"The Clinical Value of Alpha-diversity metrics to establish Dysbiosis in Microbiome Studies","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-21 18:12:33","doi":"10.21203/rs.3.rs-7987343/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2025-11-11T14:33:01+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-11T14:29:24+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-11-10T05:07:57+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-06T10:39:46+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-11-06T10:23:18+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b244d9f3-ef6e-4e13-b013-356d34f3601b","owner":[],"postedDate":"November 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":58093877,"name":"Health sciences/Biomarkers"},{"id":58093878,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":58093879,"name":"Health sciences/Diseases"},{"id":58093880,"name":"Health sciences/Gastroenterology"},{"id":58093881,"name":"Health sciences/Health care"},{"id":58093882,"name":"Health sciences/Medical research"},{"id":58093883,"name":"Biological sciences/Microbiology"}],"tags":[],"updatedAt":"2025-11-21T18:12:33+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-21 18:12:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7987343","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7987343","identity":"rs-7987343","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}