Gut-microbiota in Colorectal Cancer Patients: 16S rRNA Sequencing analysis and Machine-learning Algorithm Prediction

Gut-microbiota in Colorectal Cancer Patients: 16S rRNA Sequencing analysis and Machine-learning Algorithm Prediction | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Gut-microbiota in Colorectal Cancer Patients: 16S rRNA Sequencing analysis and Machine-learning Algorithm Prediction Yang Li, Zhaochu Wang, Chenzi Zhao, Jing Wang, Xun Wang, Linzhen Jiang, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6286866/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Objective: The incidence and mortality of colorectal cancer (CRC) have been increasing, making research into factors related to CRC necessary. This article aims to differentiate characteristics of gut microbiota between CRC patients and healthy individuals, and employs machine-learning algorithms for predicting specific microbial taxa. Methods: We conducted a multicenter case-control study starting in 2020, used 16S rRNA gene sequencing to analyze the gut microbiota in newly diagnosed CRC patients and healthy individuals. We used Python (version 3.9) to develop predictive models based on machine-learning algorithms. Results : Our research indicates a significant abundance of Escherichia-Shigella and Bacteroides in CRC patients, while Blautia and Faecalibacterium notably increased in healthy individuals. Using the Lasso model, we identified eight specific microbial taxa associated with CRC patients and thirteen taxa associated with healthy individuals. Discussion : The research highlights significant increase of various microbial taxa associated among CRC patients and healthy individuals, and also some microbiota with contentious functionalities. Among the machine-learning algorithms tested, the Random Forest model proved most suitable for predictive modeling in this region. 16S rRNA colorectal cancer machine-learning algorithm Gut microbiota Figures Figure 1 Figure 2 Figure 3 Figure 4 1. Introduction Colorectal cancer (CRC) is one of the three most common cancers and ranks the highest in incidence rates of malignant tumors in China, also the second leading cause of cancer-related deaths [ 1 , 2 ]. Early diagnosis of CRC is challenging because it is typically asymptomatic in its early stages. Approximately 70% of diagnosed patients in China present at an advanced stage, characterized by symptoms such as abdominal distension, indigestion, changes in bowel habits, abdominal pain, and bloody stools [ 3 ]. These symptoms are mostly nonspecific in early diagnosis. Even with standard treatments, up to 30% of stage II/III CRC patients experience recurrence and metastasis [ 4 ]. The gut harbors a diverse microbiota including bacteria, fungi, viruses, and other microorganisms, collectively known as the gut microbiota [ 5 , 6 ]. These microorganisms play crucial roles in human health, such as aiding digestion, maintaining intestinal mucosal barrier function, and modulating the immune system [ 7 , 8 ]. The gut microbiota is closely associated with CRC, which often arises from colorectal adenomas or chronic inflammation. An increasing epidemiological evidence supports the presence of gut microbiota dysbiosis in CRC patients [ 9 ]. The gut microbiota may promote the onset and progression of CRC through mechanisms such as specific pathogenic bacteria, influencing genetic or epigenetic changes, inducing immune dysregulation and pro-inflammatory states, altering dietary metabolism, and producing genotoxins [ 10 , 11 ]. This study used 16S rRNA sequencing to examine the gut microbiota in newly diagnosed CRC patients. Our aim was to preliminarily explore the distribution characteristics of the gut microbiota in CRC patients in this region and utilize machine-learning algorithms to construct models for predicting CRC-associated gut microbiota. 2. Materials and Methods 2.1 Study Subjects From October 2020 to June 2023, a total of 34 participants were recruited from two hospitals in Xining, Qinghai Province, China: the Qinghai Provincial Hospital of Traditional Chinese Medicine and the Qinghai Provincial People's Hospital. Seventeen colorectal cancer patients were diagnosed and compared with 17 healthy individuals based on age and gender. This study has passed the ethical review in the Qinghai Provincial Hospital of Traditional Chinese Medicine. 2.2 16S rRNA Sequencing Fresh fecal samples were collected from patients, fully suspended in sterile PBS, and centrifuged for 5 minutes to collect the supernatant. The pellet was washed three times with sterile PBS, the supernatant discarded, and the pellet collected. The bacterial cells were fully suspended and washed with sterile PBS by vortex mixing, then aliquoted into 1.5 mL centrifuge tubes. Total genomic DNA was extracted from fecal samples using the MN NucleoSpin 96 Soil Microbial DNA extraction kit. Samples were diluted to 1 ng/µL with DEPC water and PCR amplified. PCR products were purified and quantified using Nanodrop 2000. High-throughput sequencing of bacterial rRNA genes in purified composite samples was performed by Biomarker Technologies Corporation. Sequencing results were analyzed for alpha and beta diversity, differential analysis, and other studies. 2.3 Machine-Learning Algorithm Model We use Python (version 3.9) to develop a machine learning prediction model. Given the presence of numerous features, we applied the Least Absolute Shrinkage and Selection Operator (Lasso) for feature selection, implemented using the scikit-learn library (version 1.32.4) in Python. 3. Results 3.1 Characteristics of participants We recruited 34 volunteers, including 22 males and 12 females, aged 52.32 ± 8.17 years, with the majority being Han Chinese. This comparison result shows that the characteristic comparability between patients and healthy participants (Table 1 ). Table 1 Demographic characteristics of participants (n = 17) Item Healthy participants Patients p (n = 17) (n = 17) Age [mean (SD)] 52.59 (8.39) 52.06 (8.19) 0.8 Gender(%) Male 11 (64.7) 11 (64.7) Female 6 (35.2) 6 (35.2) Ethnic group (%) 0.5 Han 12 (70.6) 11 (64.7) Tibetans 5 (29.4) 6 (35.3) Marital status(%) 0.5 Unmarried/widowed/divorced/separated 0 (0) 1 (5.8) Married 17 (100) 16 (94.2) Education (%) 0.5 < 9 years of schooling 9 (52.9) 10 (58.8) ≥ 9 years of schooling 8 (47.1) 7 (41.2) Occupational labor intensity (%) 0.4 Light 10 (58.8) 7 (41.2) Moderate 6 (35.2) 7 (41.2) Heavy 1 (5.8) 3 (17.6) BMI [mean (SD)] 23.72 (2.58) 23.81 (2.61) 0.9 3.2 Composition of Microbial Species in Cancerous and Healthy Groups We first examined the abundance of microbial taxa at the phylum and genus levels in the two groups. At the phylum level, the dominant phyla were Firmicutes, Proteobacteria, Bacteroidota, and Verrucomicrobia (Fig. 1 A). Combining the relative abundances across both groups, Firmicutes were significantly decreased, and Proteobacteria were significantly increased in the cancerous group at the phylum level (Fig. 1 B). At the genus level, the most abundant bacterial species identified were Escherichia-Shigella, Bifidobacterium, Bacteroides, Blautia, and Faecalibacterium (Fig. 1 C). These microbiota showed significantly higher abundances compared to others at the genus level. Specifically, Escherichia-Shigella and Bacteroides exhibited higher abundances in cancerous patients, whereas Blautia and Faecalibacterium were more abundant in healthy individuals (Fig. 1 D). 3.3 Alpha Diversity and Beta Diversity Analysis Further exploration of alpha diversity in the two groups was conducted by using the Chao1 index to estimate species richness. The Chao1 index was relatively lower in the cancerous group, although a Wilcoxon test revealed no statistical difference between the two groups (Fig. 2 A). The Shannon index, which measures both diversity and evenness of the microbial community, also indicated lower levels in the cancerous group; although a Wilcoxon test revealed no statistical between the two groups (Fig. 2 B). Thus, there was no significant difference in alpha diversity between the two groups. Principal Coordinate Analysis (PCoA) was used to explore beta diversity between the cancerous and healthy groups. Each point in the PCoA plot represents a sample, with its position indicating its score in the principal components analysis—a method used to reduce data dimensions and reveal differences between samples. The x-axis represents the first principal component (PCo1), accounting for 17.4% of the data variation, and the y-axis represents the second principal component (PCo2), accounting for 10.9% of the data variation. Healthy group samples are represented in green dots, while cancerous group samples are represented in orange triangles. The distributions of both groups are illustrated using elliptical confidence intervals, showing the clustering tendencies of each group on these principal components. Healthy samples are predominantly clustered on the left side of the PCoA plot, whereas cancerous samples are mainly clustered on the right side, indicating distinct differences (Fig. 2 C). Further calculation of the Bray-Curtis dissimilarity distance and a Wilcoxon statistical test demonstrated significant beta diversity differences between the two groups (Fig. 2 D), indicating substantial differences in microbial community structure between the groups. 3.4 Analysis of Microbial Differences Using Lefse analysis, we examined the abundance differences of microbial taxa at the phylum and genus levels between the two groups, with an LDA score threshold of 2 for distinguishing differential genera. At the phylum level, Proteobacteria and Fusobacteriota were abundant in the cancerous group, while Firmicutes were abundant in the healthy group (Fig. 3 A). At the genus level, Solobacterium, Hungatella, Enterorhabdus, Parvimonas, Gemella, Fusobacterium, Peptostreptococcus, Porphyromonas, and unclassified Enterobacteriaceae were found abundant in the cancer group (Fig. 3 B). Further taxonomic analysis reveals that within the Firmicutes, Clostridia, Lachnospirales, Lachnospiraceae, Eubacterium hallii group, and Dorea were predominantly abundant in the healthy group. Conversely, within Proteobacteria, Gammaproteobacteria, Enterobacterales, and Enterobacteriaceae, including Escherichia-Shigella, were abundant in the cancerous group. Meanwhile, Eubacterium hallii group, Dorea, Subdoligranulum, Romboutsia, Anaerostipes, norank_f_Ruminococcaceae, Monglobus, Ruminococcus gauvreauii group, Aldercruetzia, and Marvinbryantia were abundant in the healthy group (Fig. 3 C). 3.5 Machin- learning Algorithms and Prediction In the fields of machine learning and modern statistics, Lasso has become a popular method for feature selection due to it can produce sparse solutions during model fitting. By adding an L1 penalty term to the objective function, Lasso performs feature selection and regularization simultaneously, effectively shrinking the coefficients of less important features to zero. The regularization parameter λ controls the strength of the penalty. As λ increases, more coefficients tend towards zero, indicating that the corresponding features are excluded from the model, thus achieving feature selection. Finally we identified eight variables for modeling to mitigate overfitting. Figure 4 A and 4 B illustrate the variables selected using Lasso regression (optimal lambda value: λ = 2.49041445120218). Based on the features selected by Lasso, we employed various common machine learning classification algorithms, including Logistic Regression (Logit), K-Nearest Neighbors (KNN), Catboost, Decision Tree (DT), Random Forest (RF), XGBoost, LightGBM, AdaBoost, Multi-layer Perceptron (MLP), and Gaussian Naive Bayes (GNB). The performance of each model was evaluated by plotting ROC curves and calculating AUC values. ROC curves illustrate the relationship between True Positive Rate (TPR) and False Positive Rate (FPR), providing an intuitive tool for assessing classifier performance. Ideally, a well-performing classifier's ROC curve will be closer to the top-left corner, with AUC values closer to 1 indicating superior performance. The Random Forest (RF) model exhibited the best performance with an AUC value of 0.829, surpassing other models. The study results demonstrate that the RF model excels in distinguishing CRC from other conditions, showcasing outstanding capability. Additionally, we used the selected specific bacterial genera to establish machine learning models for predicting various diseases. In predicting cancer, the Random Forest (RF) model also performed exceptionally well with an AUC value of 0.829, outperforming other models (Fig. 4 C). Finally, based on the Lasso model, we screened eight specific microbiota associated with CRC patients and thirteen specific microbiota associated with healthy individuals (Table 2 ). Table 2 Specific bacterial genera in CRC CRC Group Healthy Group Fusobacterium Solobacterium Peptostreptococcus Hungatella Porphyromonas Enterorhabdus Escherichia.Shigella Parvimonas unclassified_f Enterobacteriaceae Gemella Enterococcus Fusobacterium Granulicatella Porphyromonas Escherichia-Shigella Escherichia-Shigella unclassified_f Enterobacteriaceae Flavonifractor norank_f Oscillospiraceae Enterococcus Granulicatella 4. Discussion The correlation between gut microbiota and the occurrence and development of colorectal cancer has drawn increasing attention. However, determining how exactly microorganisms influence cancer and its progression remains challenging. Cross-sectional epidemiological studies, microbiota analyses of fecal and colorectal tissue samples, and data predicted by computational algorithm models have revealed specific taxonomic and bacterial factors. The role of gut microbiota in the host also influences the differentiation and malignancy of CRC during the cancer process. In this case-control study, we investigated gut microbiota dysbiosis in CRC patients. Our research indicates significant differences in the abundance of gut microbiota between CRC patients and healthy individuals. Specifically, at the genus level, there is a higher abundance of Escherichia-Shigella and Bacteroides in CRC patients compared to healthy individuals. The elevation of Escherichia-Shigella and Bacteroides has been previously reported in some studies related to CRC [ 12 – 14 ]. Escherichia-Shigella not only exhibits differential abundance between CRC patients and healthy individuals but also between proximal and distal cancer sites. As a potential oncogenic pathogen, it may serve as a driving bacterium promoting tumor initiation [ 15 ]. Additionally, the genus Bacteroides has been found abundant and highly expressed in CRC patients in previous studies [ 16 ]. In contrast, the genera Blautia and Faecalibacterium exhibit higher relative abundances in healthy individuals compared to CRC patients. These bacteria are essential for maintaining the normal ecological balance of the human gut [ 17 , 18 ]. Faecalibacterium, considered a probiotic in current research, may contribute to immune evasion of tumors due to its relatively low abundance in CRC[ 19 , 20 ]. These findings highlight the different composition of gut microbiota between CRC patients and healthy individuals. Subsequently, we conducted analyses of alpha and beta diversity. Our study revealed significant alterations in the beta diversity of gut microbiota in CRC patients. However, alpha diversity analysis showed that although the Chao1 and Shannon indices were relatively lower in CRC patients, Wilcoxon tests indicated no statistically significant differences between the two groups. A previous study reported similar findings [ 21 ]. The possible reason for this could be the substantial geographic and ethnic-specific heterogeneity in gut microbiota [ 22 ]. Additionally, a notable finding in this study was the decreased abundance of Firmicutes in CRC patients. Firmicutes comprise a large bacterial phylum. A study examining 294 matched tumor and adjacent biopsy samples reported an increased abundance of Fusobacterium, Streptococcus, and Bacteroides in tumors, while Firmicutes were found decreasing [ 23 ]. The tumor microbiome also varies according to anatomical sub-sites [ 24 ].Importantly, microbial communities within individual tumors are evenly distributed and not confined to specific regions; different taxa may play distinct roles at varying distances from the tumor [ 25 ]. Our study did not consider the potential role of gut bacterial metabolites in CRC, which could be a limitation. Additionally, gut microbiota are influenced by factors such as dietary habits and medication use, which inevitably impact the findings of this analysis. Machine learning has been demonstrated as an effective method for analyzing microbiota data and predicting specific outcomes, including human and environmental health. Machine learning based on microbiota data has successfully been used to predict disease states in human health, assess environmental quality, detect pollutants in the environment, and serve as trace evidence in forensic science. Machine-learning algorithms can effectively model and predict bacteria associated with CRC, and numerous studies have already applied this approach [ 26 – 28 ]. However, evidence regarding the selection of various classification algorithms remains limited. In our study, based on Lasso-selected feature variables, we employed several common machine learning classification algorithms, including logistic regression (Logit), k-nearest neighbors (KNN), and random forest (RF), among others. Among these, the random forest (RF) model exhibited the most outstanding performance with an AUC value of 0.829. Its application has been validated in several studies [ 29 – 31 ]. Finally, we identified predictive genera selected by the Lasso model, including Fusobacterium, Peptostreptococcus, Porphyromonas, Escherichia-Shigella, Enterococcus, and Granulicatella, which have been reported to potentially correlate with CRC [ 32 – 37 ]. We employed machine-learning algorithms to establish predictive models tailored to CRC-associated gut microbiota, aiming to enhance diagnostic accuracy for pathogenic gut microbiota. However, our study has limitations, particularly the need for larger sample sizes to strengthen our findings. Future research may benefit from longitudinal studies to further elucidate the dynamic role of gut microbiota in the pathogenesis of CRC. 5. Conclusions Our study conducted a preliminary exploration of the gut microbiota composition in colorectal cancer (CRC) patients. A significant abundance of Escherichia-Shigella and Bacteroides was observed in CRC patients, whereas Blautia and Faecalibacterium were notably increased in healthy individuals. Using the Lasso model, we identified eight specific microbial taxa associated with CRC patients and thirteen taxa associated with healthy individuals. These findings contribute to a better understanding of the gut microbiota characteristics in CRC patients, and provide a reference for further research. Declarations Ethics approval and consent to participate ： In the present study, we adhered to the research standards set forth by the Declaration of Helsinki, established by the World Medical Association. The study protocol was reviewed and approved by the Ethics Committee of Qinghai Provincial Hospital of Traditional Chinese Medicine. Participation in the study was contingent upon the provision of permission by all patients involved. To maintain confidentiality, patient IDs were removed after the collection of clinicopathological data, thereby preventing direct or indirect identifcation of patients. Consent for publication: All authors agree to submit the article for publication. Author Contributions: X.J.W. designed the research. Y.L. and Z.W. wrote the manuscript. C.Z. and X.W. performed the 16sRNA sequencing and data collection. S.T., L.J. and J.W. analyzed the data. X.J.W. and R.S. provided critical revisions and conceptual insights. All authors contributed to the article and approved the submitted version. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by the Natural Science Foundation of Qinghai Province (2020-ZJ-768) and Clinical Research Center for Traditional Chinese Medicine on Anorectal and Perianal Wound Repair of Fujian Province (2022Y2011). Data Availability Statement: The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors. Acknowledgments: We express our profound appreciation to all the volunteers who contributed to this study. Their collective effort and unwavering dedication were indispensable to the successful completion of our research. We would also like to thank our colleagues in the hospital who participated in this project. Conflicts of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher’s note : All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors, and the reviewers. Any product that may be evaluated in this article, or any claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. References Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin. 2024;74(1):12-49. Qiu H, Cao S, Xu R. Cancer incidence, mortality, and burden in China: a time-trend analysis and comparison with the United States and United Kingdom based on the global epidemiological data released in 2020. Cancer Commun (Lond). 2021;41(10):1037-48. Zheng RS, Chen R, Han BF, Wang SM, Li L, Sun KX, et al. [Cancer incidence and mortality in China, 2022]. Zhonghua Zhong Liu Za Zhi. 2024;46(3):221-31. Jiang X, Yang L, Chen G, Feng X, Liu Y, Gao Q, et al. Discovery of Kinetin in inhibiting colorectal cancer progression via enhancing PSMB1-mediated RAB34 degradation. Cancer Lett. 2024;584:216600. Leclercq S, de Timary P. Role of the Microbiome and the Gut-Brain Axis in Alcohol Use Disorder: Potential Implication for Treatment Development. Curr Top Behav Neurosci. 2024. Gorny RL, Golofit-Szymczak M, Pawlak A, Lawniczek-Walczyk A, Cyprowski M, Stobnicka A, et al. Effectiveness of UV-C radiation in inactivation of microorganisms on materials with different surface structures. Ann Agric Environ Med. 2024;31(2):287-93. Wijesekara T, Abeyrathne E, Ahn DU. Effect of Bioactive Peptides on Gut Microbiota and Their Relations to Human Health. Foods. 2024;13(12). Baffy G, Portincasa P. Gut Microbiota and Sinusoidal Vasoregulation in MASLD: A Portal Perspective. Metabolites. 2024;14(6). Abdullah M, Sukartini N, Nursyirwan SA, Pribadi RR, Maulahela H, Utari AP, et al. Gut Microbiota Profiles in Early- and Late-Onset Colorectal Cancer: A Potential Diagnostic Biomarker in the Future. Digestion. 2021;102(6):823-32. Hu YZ, Wu CS, Wang J, Han XQ, Si PY, Zhang YA, et al. Antimicrobial Protein LECT2-b Helps Maintain Gut Microbiota Homeostasis via Selectively Targeting Certain Pathogenic Bacteria. J Immunol. 2024;212(1):81-95. Lee C, Lee S, Yoo W. Metabolic Interaction Between Host and the Gut Microbiota During High-Fat Diet-Induced Colorectal Cancer. J Microbiol. 2024;62(3):153-65. Zhang H, Lan M, Cui G, Zhu W. The Influence of Caerulomycin A on the Intestinal Microbiota in SD Rats. Mar Drugs. 2020;18(5). Tajasuwan L, Kettawan A, Rungruang T, Wunjuntuk K, Prombutara P. Role of Dietary Defatted Rice Bran in the Modulation of Gut Microbiota in AOM/DSS-Induced Colitis-Associated Colorectal Cancer Rat Model. Nutrients. 2023;15(6). Avuthu N, Guda C. Meta-Analysis of Altered Gut Microbiota Reveals Microbial and Metabolic Biomarkers for Colorectal Cancer. Microbiol Spectr. 2022;10(4):e0001322. Gao Z, Guo B, Gao R, Zhu Q, Qin H. Microbiota disbiosis is associated with colorectal cancer. Front Microbiol. 2015;6:20. Huh JW, Kim MJ, Kim J, Lee HG, Ryoo SB, Ku JL, et al. Enterotypical Prevotella and three novel bacterial biomarkers in preoperative stool predict the clinical outcome of colorectal cancer. Microbiome. 2022;10(1):203. Liu G, Li T, Zhu X, Zhang X, Wang J. An independent evaluation in a CRC patient cohort of microbiome 16S rRNA sequence analysis methods: OTU clustering, DADA2, and Deblur. Front Microbiol. 2023;14:1178744. Jin M, Wu J, Shi L, Zhou B, Shang F, Chang X, et al. Gut microbiota distinct between colorectal cancers with deficient and proficient mismatch repair: A study of 230 CRC patients. Front Microbiol. 2022;13:993285. Martin R, Rios-Covian D, Huillet E, Auger S, Khazaal S, Bermudez-Humaran LG, et al. Faecalibacterium: a bacterial genus with promising human health applications. FEMS Microbiol Rev. 2023;47(4). Gopalakrishnan V, Spencer CN, Nezi L, Reuben A, Andrews MC, Karpinets TV, et al. Gut microbiome modulates response to anti-PD-1 immunotherapy in melanoma patients. Science. 2018;359(6371):97-103. Dai Z, Coker OO, Nakatsu G, Wu WKK, Zhao L, Chen Z, et al. Multi-cohort analysis of colorectal cancer metagenome identified altered bacteria across populations and universal bacterial markers. Microbiome. 2018;6(1):70. Wong CC, Yu J. Gut microbiota in colorectal cancer development and therapy. Nat Rev Clin Oncol. 2023;20(7):429-52. Shah MS, DeSantis T, Yamal JM, Weir T, Ryan EP, Cope JL, et al. Re-purposing 16S rRNA gene sequence data from within case paired tumor biopsy and tumor-adjacent biopsy or fecal samples to identify microbial markers for colorectal cancer. PLoS One. 2018;13(11):e0207002. Omar Al-Hassi H, Ng O, Brookes M. Tumour-associated and non-tumour-associated microbiota in colorectal cancer. Gut. 2018;67(2):395. Murphy CL, Barrett M, Pellanda P, Killeen S, McCourt M, Andrews E, et al. Mapping the colorectal tumor microbiota. Gut Microbes. 2021;13(1):1-10. Qi Z, Zhibo Z, Jing Z, Zhanbo Q, Shugao H, Weili J, et al. Prediction model of poorly differentiated colorectal cancer (CRC) based on gut bacteria. BMC Microbiol. 2022;22(1):312. Huang Z, Huang X, Huang Y, Liang K, Chen L, Zhong C, et al. Identification of KRAS mutation-associated gut microbiota in colorectal cancer and construction of predictive machine learning model. Microbiol Spectr. 2024;12(5):e0272023. Konishi Y, Okumura S, Matsumoto T, Itatani Y, Nishiyama T, Okazaki Y, et al. Development and evaluation of a colorectal cancer screening method using machine learning-based gut microbiota analysis. Cancer Med. 2022;11(16):3194-206. Liu J, Huang X, Chen C, Wang Z, Huang Z, Qin M, et al. Identification of colorectal cancer progression-associated intestinal microbiome and predictive signature construction. J Transl Med. 2023;21(1):373. Luu K, Ye JY, Lagishetty V, Liang F, Hauer M, Sedighian F, et al. Fecal and Tissue Microbiota Are Associated with Tumor T-Cell Infiltration and Mesenteric Lymph Node Involvement in Colorectal Cancer. Nutrients. 2023;15(2). Wu Y, Jiao N, Zhu R, Zhang Y, Wu D, Wang AJ, et al. Identification of microbial markers across populations in early detection of colorectal cancer. Nat Commun. 2021;12(1):3063. Chen S, Zhang L, Li M, Zhang Y, Sun M, Wang L, et al. Fusobacterium nucleatum reduces METTL3-mediated m(6)A modification and contributes to colorectal cancer metastasis. Nat Commun. 2022;13(1):1248. Cui W, Guo M, Liu D, Xiao P, Yang C, Huang H, et al. Gut microbial metabolite facilitates colorectal cancer development via ferroptosis inhibition. Nat Cell Biol. 2024;26(1):124-37. Kerdreux M, Edin S, Lowenmark T, Bronnec V, Lofgren-Burstrom A, Zingmark C, et al. Porphyromonas gingivalis in Colorectal Cancer and its Association to Patient Prognosis. J Cancer. 2023;14(9):1479-85. Zhong X, Wang Y, Xu J, Cao H, Zhang F, Wang X. Gut microbiota signatures in tissues of the colorectal polyp and normal colorectal mucosa, and faeces. Front Cell Infect Microbiol. 2022;12:1054808. Elahi Z, Shariati A, Bostanghadiri N, Dadgar-Zankbar L, Razavi S, Norzaee S, et al. Association of Lactobacillus, Firmicutes, Bifidobacterium, Clostridium, and Enterococcus with colorectal cancer in Iranian patients. Heliyon. 2023;9(12):e22602. Alexander JL, Posma JM, Scott A, Poynter L, Mason SE, Doria ML, et al. Pathobionts in the tumour microbiota predict survival following resection for colorectal cancer. Microbiome. 2023;11(1):100. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-6286866","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":437415803,"identity":"0112583b-5b81-486e-976b-90c218ce5611","order_by":0,"name":"Yang Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvklEQVRIiWNgGAWjYBACNmb+jw8SKmzq+9kbiNTCx85gbPDhTBrjzJ4DRGqR42cwk5zZdphxw40Eoh3GkGzMc+Yws8HNxxtvMNTYRBOj5eBjnop0NsnbacUWDMfSchsIa2FsBtpizcN3O8dMgrHhMDFamNmkeduYJRhuniFaCxsb0PvOBgI3eIjWwsMMCuQEyR6gXxKI8Yt8/xlGUFQm8LMf3njjQ40NYS3IwEAigRTlEC2k6hgFo2AUjIKRAQDERjtvfTQgPwAAAABJRU5ErkJggg==","orcid":"","institution":"The Affiliated People’s Hospital of Fujian University of Traditional Chinese Medicine","correspondingAuthor":true,"prefix":"","firstName":"Yang","middleName":"","lastName":"Li","suffix":""},{"id":437415804,"identity":"5dcbad93-65e8-42e1-a398-6f4ead7da90e","order_by":1,"name":"Zhaochu Wang","email":"","orcid":"","institution":"The Affiliated People’s Hospital of Fujian University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Zhaochu","middleName":"","lastName":"Wang","suffix":""},{"id":437415805,"identity":"664c1df8-e0bc-4a4f-ba05-7a3410c69961","order_by":2,"name":"Chenzi Zhao","email":"","orcid":"","institution":"The Affiliated People’s Hospital of Fujian University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Chenzi","middleName":"","lastName":"Zhao","suffix":""},{"id":437415806,"identity":"11a7342e-0b4d-48c2-9148-ea11c63ffa45","order_by":3,"name":"Jing Wang","email":"","orcid":"","institution":"The Affiliated People’s Hospital of Fujian University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Wang","suffix":""},{"id":437415807,"identity":"9cd669d9-7d5b-4a73-8c07-9f4cc4d6d2f3","order_by":4,"name":"Xun Wang","email":"","orcid":"","institution":"The Affiliated People’s Hospital of Fujian University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Xun","middleName":"","lastName":"Wang","suffix":""},{"id":437415808,"identity":"020bc1c7-d62b-4e00-acef-4c6abfdaf77e","order_by":5,"name":"Linzhen Jiang","email":"","orcid":"","institution":"The Affiliated People’s Hospital of Fujian University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Linzhen","middleName":"","lastName":"Jiang","suffix":""},{"id":437415809,"identity":"715f20c6-6bd8-42b0-8b11-8a01abb20aa1","order_by":6,"name":"Jiemin Huang","email":"","orcid":"","institution":"The Affiliated People’s Hospital of Fujian University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jiemin","middleName":"","lastName":"Huang","suffix":""},{"id":437415810,"identity":"0af24a19-6bdf-449c-b5b9-70540ff7ad43","order_by":7,"name":"Shou Tian","email":"","orcid":"","institution":"Medical College of Qinghai University","correspondingAuthor":false,"prefix":"","firstName":"Shou","middleName":"","lastName":"Tian","suffix":""},{"id":437415811,"identity":"de5c80d8-6d55-40c1-b2e9-93eb4fb0aeb0","order_by":8,"name":"Xiaojin Wang","email":"","orcid":"","institution":"Provincial Hospital of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Xiaojin","middleName":"","lastName":"Wang","suffix":""},{"id":437415812,"identity":"6eae8c20-489a-46cb-9cab-a5f73edf2019","order_by":9,"name":"Rong Shi","email":"","orcid":"","institution":"The Affiliated People’s Hospital of Fujian University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Rong","middleName":"","lastName":"Shi","suffix":""}],"badges":[],"createdAt":"2025-03-23 07:08:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6286866/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6286866/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":80045770,"identity":"182e6b18-2616-4b31-9bec-62a45c888868","added_by":"auto","created_at":"2025-04-07 09:45:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1148081,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of the abundance of gut microbiota in patients. (A) Stacking diagram of microbial species in the cancerous group and healthy group at the genus level of each sample. (B) Horizontal microbial species box diagram between two groups. (C) Stacking diagram of microbial species in cancerous group and healthy group at the genus level of each sample (D) Box plot of microbial species at the phylum level between two groups. CTL and CA were for healthy individuals and CRC patients, respectively.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6286866/v1/d0d44ff7451ed500dc1ecfff.png"},{"id":80047222,"identity":"5c995e36-2608-4218-a86c-c5059939f783","added_by":"auto","created_at":"2025-04-07 09:53:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":302217,"visible":true,"origin":"","legend":"\u003cp\u003eDiversity analysis of gut microbiota in CRC patients and healthy individuals. (A) The box plot of the Chao1 index between the two groups. (B) The box plot of Shannon index between the two groups. (C) PCoa scatter plot analysis of differences in gut microbiota between the two. (D) Bray Curtis distance box plot.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6286866/v1/d61fdd488d8c00693caf1198.png"},{"id":80045766,"identity":"27ee2c11-9e57-4028-9cbe-d9f61e73e841","added_by":"auto","created_at":"2025-04-07 09:45:51","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":558767,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of differences in gut microbiota between CRC patients and healthy individuals. (A) Lefse analysis at the gate level (B) Lefse analysis at the genus level (C) Lefse analysis at multiple levels.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6286866/v1/0618ec0aa52a631c19483341.png"},{"id":80045772,"identity":"e5c21c2a-3b17-4555-9178-8fa926b7df09","added_by":"auto","created_at":"2025-04-07 09:45:51","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":384947,"visible":true,"origin":"","legend":"\u003cp\u003eMachine learning algorithm prediction model. The (A, B) Lasso bullet plot and coefficient plot show the variables selected by Lasso regression. The (C) ROC curve screens 10 different prediction models.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6286866/v1/d35820fb991bf713c06c4212.png"},{"id":89045472,"identity":"c5f78d57-924c-4d2d-a9a0-e2337e057749","added_by":"auto","created_at":"2025-08-14 06:38:53","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2783784,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6286866/v1/b7ae133a-bfa1-4d65-8b41-ca7b52830193.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Gut-microbiota in Colorectal Cancer Patients: 16S rRNA Sequencing analysis and Machine-learning Algorithm Prediction","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eColorectal cancer (CRC) is one of the three most common cancers and ranks the highest in incidence rates of malignant tumors in China, also the second leading cause of cancer-related deaths [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Early diagnosis of CRC is challenging because it is typically asymptomatic in its early stages. Approximately 70% of diagnosed patients in China present at an advanced stage, characterized by symptoms such as abdominal distension, indigestion, changes in bowel habits, abdominal pain, and bloody stools [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. These symptoms are mostly nonspecific in early diagnosis. Even with standard treatments, up to 30% of stage II/III CRC patients experience recurrence and metastasis [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe gut harbors a diverse microbiota including bacteria, fungi, viruses, and other microorganisms, collectively known as the gut microbiota [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. These microorganisms play crucial roles in human health, such as aiding digestion, maintaining intestinal mucosal barrier function, and modulating the immune system [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The gut microbiota is closely associated with CRC, which often arises from colorectal adenomas or chronic inflammation. An increasing epidemiological evidence supports the presence of gut microbiota dysbiosis in CRC patients [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The gut microbiota may promote the onset and progression of CRC through mechanisms such as specific pathogenic bacteria, influencing genetic or epigenetic changes, inducing immune dysregulation and pro-inflammatory states, altering dietary metabolism, and producing genotoxins [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. This study used 16S rRNA sequencing to examine the gut microbiota in newly diagnosed CRC patients. Our aim was to preliminarily explore the distribution characteristics of the gut microbiota in CRC patients in this region and utilize machine-learning algorithms to construct models for predicting CRC-associated gut microbiota.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study Subjects\u003c/h2\u003e \u003cp\u003e From October 2020 to June 2023, a total of 34 participants were recruited from two hospitals in Xining, Qinghai Province, China: the Qinghai Provincial Hospital of Traditional Chinese Medicine and the Qinghai Provincial People's Hospital. Seventeen colorectal cancer patients were diagnosed and compared with 17 healthy individuals based on age and gender. This study has passed the ethical review in the Qinghai Provincial Hospital of Traditional Chinese Medicine.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 16S rRNA Sequencing\u003c/h2\u003e \u003cp\u003eFresh fecal samples were collected from patients, fully suspended in sterile PBS, and centrifuged for 5 minutes to collect the supernatant. The pellet was washed three times with sterile PBS, the supernatant discarded, and the pellet collected. The bacterial cells were fully suspended and washed with sterile PBS by vortex mixing, then aliquoted into 1.5 mL centrifuge tubes. Total genomic DNA was extracted from fecal samples using the MN NucleoSpin 96 Soil Microbial DNA extraction kit. Samples were diluted to 1 ng/\u0026micro;L with DEPC water and PCR amplified. PCR products were purified and quantified using Nanodrop 2000. High-throughput sequencing of bacterial rRNA genes in purified composite samples was performed by Biomarker Technologies Corporation. Sequencing results were analyzed for alpha and beta diversity, differential analysis, and other studies.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Machine-Learning Algorithm Model\u003c/h2\u003e \u003cp\u003eWe use Python (version 3.9) to develop a machine learning prediction model. Given the presence of numerous features, we applied the Least Absolute Shrinkage and Selection Operator (Lasso) for feature selection, implemented using the scikit-learn library (version 1.32.4) in Python.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Characteristics of participants\u003c/h2\u003e \u003cp\u003eWe recruited 34 volunteers, including 22 males and 12 females, aged 52.32\u0026thinsp;\u0026plusmn;\u0026thinsp;8.17 years, with the majority being Han Chinese. This comparison result shows that the characteristic comparability between patients and healthy participants (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDemographic characteristics of participants (n\u0026thinsp;=\u0026thinsp;17)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eItem\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHealthy participants\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePatients\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge [mean (SD)]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e52.59 (8.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e52.06 (8.19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11 (64.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11 (64.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6 (35.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (35.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEthnic group (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12 (70.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11 (64.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTibetans\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5 (29.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (35.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMarital status(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnmarried/widowed/divorced/separated\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1 (5.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMarried\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e17 (100)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16 (94.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;9 years of schooling\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9 (52.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10 (58.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026ge;\u0026thinsp;9 years of schooling\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8 (47.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (41.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOccupational labor intensity (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10 (58.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (41.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModerate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6 (35.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (41.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeavy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1 (5.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (17.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI [mean (SD)]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23.72 (2.58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.81 (2.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Composition of Microbial Species in Cancerous and Healthy Groups\u003c/h2\u003e \u003cp\u003eWe first examined the abundance of microbial taxa at the phylum and genus levels in the two groups. At the phylum level, the dominant phyla were Firmicutes, Proteobacteria, Bacteroidota, and Verrucomicrobia (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). Combining the relative abundances across both groups, Firmicutes were significantly decreased, and Proteobacteria were significantly increased in the cancerous group at the phylum level (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003eAt the genus level, the most abundant bacterial species identified were Escherichia-Shigella, Bifidobacterium, Bacteroides, Blautia, and Faecalibacterium (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). These microbiota showed significantly higher abundances compared to others at the genus level. Specifically, Escherichia-Shigella and Bacteroides exhibited higher abundances in cancerous patients, whereas Blautia and Faecalibacterium were more abundant in healthy individuals (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Alpha Diversity and Beta Diversity Analysis\u003c/h2\u003e \u003cp\u003eFurther exploration of alpha diversity in the two groups was conducted by using the Chao1 index to estimate species richness. The Chao1 index was relatively lower in the cancerous group, although a Wilcoxon test revealed no statistical difference between the two groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). The Shannon index, which measures both diversity and evenness of the microbial community, also indicated lower levels in the cancerous group; although a Wilcoxon test revealed no statistical between the two groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Thus, there was no significant difference in alpha diversity between the two groups.\u003c/p\u003e \u003cp\u003ePrincipal Coordinate Analysis (PCoA) was used to explore beta diversity between the cancerous and healthy groups. Each point in the PCoA plot represents a sample, with its position indicating its score in the principal components analysis\u0026mdash;a method used to reduce data dimensions and reveal differences between samples. The x-axis represents the first principal component (PCo1), accounting for 17.4% of the data variation, and the y-axis represents the second principal component (PCo2), accounting for 10.9% of the data variation. Healthy group samples are represented in green dots, while cancerous group samples are represented in orange triangles. The distributions of both groups are illustrated using elliptical confidence intervals, showing the clustering tendencies of each group on these principal components. Healthy samples are predominantly clustered on the left side of the PCoA plot, whereas cancerous samples are mainly clustered on the right side, indicating distinct differences (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Further calculation of the Bray-Curtis dissimilarity distance and a Wilcoxon statistical test demonstrated significant beta diversity differences between the two groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), indicating substantial differences in microbial community structure between the groups.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Analysis of Microbial Differences\u003c/h2\u003e \u003cp\u003eUsing Lefse analysis, we examined the abundance differences of microbial taxa at the phylum and genus levels between the two groups, with an LDA score threshold of 2 for distinguishing differential genera. At the phylum level, Proteobacteria and Fusobacteriota were abundant in the cancerous group, while Firmicutes were abundant in the healthy group (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). At the genus level, Solobacterium, Hungatella, Enterorhabdus, Parvimonas, Gemella, Fusobacterium, Peptostreptococcus, Porphyromonas, and unclassified Enterobacteriaceae were found abundant in the cancer group (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003eFurther taxonomic analysis reveals that within the Firmicutes, Clostridia, Lachnospirales, Lachnospiraceae, Eubacterium hallii group, and Dorea were predominantly abundant in the healthy group. Conversely, within Proteobacteria, Gammaproteobacteria, Enterobacterales, and Enterobacteriaceae, including Escherichia-Shigella, were abundant in the cancerous group. Meanwhile, Eubacterium hallii group, Dorea, Subdoligranulum, Romboutsia, Anaerostipes, norank_f_Ruminococcaceae, Monglobus, Ruminococcus gauvreauii group, Aldercruetzia, and Marvinbryantia were abundant in the healthy group (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Machin- learning Algorithms and Prediction\u003c/h2\u003e \u003cp\u003eIn the fields of machine learning and modern statistics, Lasso has become a popular method for feature selection due to it can produce sparse solutions during model fitting. By adding an L1 penalty term to the objective function, Lasso performs feature selection and regularization simultaneously, effectively shrinking the coefficients of less important features to zero. The regularization parameter λ controls the strength of the penalty. As λ increases, more coefficients tend towards zero, indicating that the corresponding features are excluded from the model, thus achieving feature selection. Finally we identified eight variables for modeling to mitigate overfitting. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB illustrate the variables selected using Lasso regression (optimal lambda value: λ\u0026thinsp;=\u0026thinsp;2.49041445120218).\u003c/p\u003e \u003cp\u003eBased on the features selected by Lasso, we employed various common machine learning classification algorithms, including Logistic Regression (Logit), K-Nearest Neighbors (KNN), Catboost, Decision Tree (DT), Random Forest (RF), XGBoost, LightGBM, AdaBoost, Multi-layer Perceptron (MLP), and Gaussian Naive Bayes (GNB). The performance of each model was evaluated by plotting ROC curves and calculating AUC values. ROC curves illustrate the relationship between True Positive Rate (TPR) and False Positive Rate (FPR), providing an intuitive tool for assessing classifier performance. Ideally, a well-performing classifier's ROC curve will be closer to the top-left corner, with AUC values closer to 1 indicating superior performance.\u003c/p\u003e \u003cp\u003eThe Random Forest (RF) model exhibited the best performance with an AUC value of 0.829, surpassing other models. The study results demonstrate that the RF model excels in distinguishing CRC from other conditions, showcasing outstanding capability. Additionally, we used the selected specific bacterial genera to establish machine learning models for predicting various diseases. In predicting cancer, the Random Forest (RF) model also performed exceptionally well with an AUC value of 0.829, outperforming other models (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). Finally, based on the Lasso model, we screened eight specific microbiota associated with CRC patients and thirteen specific microbiota associated with healthy individuals (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSpecific bacterial genera in CRC\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRC Group\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHealthy Group\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFusobacterium\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSolobacterium\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePeptostreptococcus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHungatella\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePorphyromonas\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEnterorhabdus\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEscherichia.Shigella\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eParvimonas\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eunclassified_f Enterobacteriaceae\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGemella\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEnterococcus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFusobacterium\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGranulicatella\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePorphyromonas\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEscherichia-Shigella\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEscherichia-Shigella\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eunclassified_f Enterobacteriaceae\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFlavonifractor\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003enorank_f Oscillospiraceae\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEnterococcus\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGranulicatella\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eThe correlation between gut microbiota and the occurrence and development of colorectal cancer has drawn increasing attention. However, determining how exactly microorganisms influence cancer and its progression remains challenging. Cross-sectional epidemiological studies, microbiota analyses of fecal and colorectal tissue samples, and data predicted by computational algorithm models have revealed specific taxonomic and bacterial factors.\u003c/p\u003e \u003cp\u003eThe role of gut microbiota in the host also influences the differentiation and malignancy of CRC during the cancer process. In this case-control study, we investigated gut microbiota dysbiosis in CRC patients. Our research indicates significant differences in the abundance of gut microbiota between CRC patients and healthy individuals. Specifically, at the genus level, there is a higher abundance of Escherichia-Shigella and Bacteroides in CRC patients compared to healthy individuals. The elevation of Escherichia-Shigella and Bacteroides has been previously reported in some studies related to CRC [\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Escherichia-Shigella not only exhibits differential abundance between CRC patients and healthy individuals but also between proximal and distal cancer sites. As a potential oncogenic pathogen, it may serve as a driving bacterium promoting tumor initiation [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Additionally, the genus Bacteroides has been found abundant and highly expressed in CRC patients in previous studies [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn contrast, the genera Blautia and Faecalibacterium exhibit higher relative abundances in healthy individuals compared to CRC patients. These bacteria are essential for maintaining the normal ecological balance of the human gut [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Faecalibacterium, considered a probiotic in current research, may contribute to immune evasion of tumors due to its relatively low abundance in CRC[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. These findings highlight the different composition of gut microbiota between CRC patients and healthy individuals.\u003c/p\u003e \u003cp\u003eSubsequently, we conducted analyses of alpha and beta diversity. Our study revealed significant alterations in the beta diversity of gut microbiota in CRC patients. However, alpha diversity analysis showed that although the Chao1 and Shannon indices were relatively lower in CRC patients, Wilcoxon tests indicated no statistically significant differences between the two groups. A previous study reported similar findings [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. The possible reason for this could be the substantial geographic and ethnic-specific heterogeneity in gut microbiota [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAdditionally, a notable finding in this study was the decreased abundance of Firmicutes in CRC patients. Firmicutes comprise a large bacterial phylum. A study examining 294 matched tumor and adjacent biopsy samples reported an increased abundance of Fusobacterium, Streptococcus, and Bacteroides in tumors, while Firmicutes were found decreasing [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. The tumor microbiome also varies according to anatomical sub-sites [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].Importantly, microbial communities within individual tumors are evenly distributed and not confined to specific regions; different taxa may play distinct roles at varying distances from the tumor [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Our study did not consider the potential role of gut bacterial metabolites in CRC, which could be a limitation. Additionally, gut microbiota are influenced by factors such as dietary habits and medication use, which inevitably impact the findings of this analysis.\u003c/p\u003e \u003cp\u003eMachine learning has been demonstrated as an effective method for analyzing microbiota data and predicting specific outcomes, including human and environmental health. Machine learning based on microbiota data has successfully been used to predict disease states in human health, assess environmental quality, detect pollutants in the environment, and serve as trace evidence in forensic science. Machine-learning algorithms can effectively model and predict bacteria associated with CRC, and numerous studies have already applied this approach [\u003cspan additionalcitationids=\"CR27\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. However, evidence regarding the selection of various classification algorithms remains limited. In our study, based on Lasso-selected feature variables, we employed several common machine learning classification algorithms, including logistic regression (Logit), k-nearest neighbors (KNN), and random forest (RF), among others. Among these, the random forest (RF) model exhibited the most outstanding performance with an AUC value of 0.829. Its application has been validated in several studies [\u003cspan additionalcitationids=\"CR30\" citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Finally, we identified predictive genera selected by the Lasso model, including Fusobacterium, Peptostreptococcus, Porphyromonas, Escherichia-Shigella, Enterococcus, and Granulicatella, which have been reported to potentially correlate with CRC [\u003cspan additionalcitationids=\"CR33 CR34 CR35 CR36\" citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWe employed machine-learning algorithms to establish predictive models tailored to CRC-associated gut microbiota, aiming to enhance diagnostic accuracy for pathogenic gut microbiota. However, our study has limitations, particularly the need for larger sample sizes to strengthen our findings. Future research may benefit from longitudinal studies to further elucidate the dynamic role of gut microbiota in the pathogenesis of CRC.\u003c/p\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eOur study conducted a preliminary exploration of the gut microbiota composition in colorectal cancer (CRC) patients. A significant abundance of Escherichia-Shigella and Bacteroides was observed in CRC patients, whereas Blautia and Faecalibacterium were notably increased in healthy individuals. Using the Lasso model, we identified eight specific microbial taxa associated with CRC patients and thirteen taxa associated with healthy individuals. These findings contribute to a better understanding of the gut microbiota characteristics in CRC patients, and provide a reference for further research.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003cstrong\u003e：\u003c/strong\u003eIn the present study, we adhered to the research standards set forth by the Declaration of Helsinki, established by the World Medical Association. The study protocol was reviewed and approved by the Ethics Committee of Qinghai Provincial Hospital of Traditional Chinese Medicine. Participation in the study was contingent upon the provision of permission by all patients involved. To maintain confidentiality, patient IDs were removed after the collection of clinicopathological data, thereby preventing direct or indirect identifcation of patients.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication:\u0026nbsp;\u003c/strong\u003eAll authors agree to submit the article for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions:\u0026nbsp;\u003c/strong\u003eX.J.W. designed the research. Y.L. and Z.W. wrote the manuscript. C.Z. and X.W. performed the 16sRNA sequencing and data collection. S.T., L.J. and J.W. analyzed the data. X.J.W. and R.S. provided critical revisions and conceptual insights. All authors contributed to the article and approved the submitted version. All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e This work was supported by the Natural Science Foundation of Qinghai Province (2020-ZJ-768) and Clinical Research Center for Traditional Chinese Medicine on Anorectal and Perianal Wound Repair of Fujian Province (2022Y2011).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement:\u003c/strong\u003e The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u003c/strong\u003e We express our profound appreciation to all the volunteers who contributed to this study. Their collective effort and unwavering dedication were indispensable to the successful completion of our research. We would also like to thank our colleagues in the hospital who participated in this project.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest:\u0026nbsp;\u003c/strong\u003eThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePublisher\u0026rsquo;s note\u003c/strong\u003e: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors, and the reviewers. Any product that may be evaluated in this article, or any claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSiegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin. 2024;74(1):12-49.\u003c/li\u003e\n\u003cli\u003eQiu H, Cao S, Xu R. Cancer incidence, mortality, and burden in China: a time-trend analysis and comparison with the United States and United Kingdom based on the global epidemiological data released in 2020. Cancer Commun (Lond). 2021;41(10):1037-48.\u003c/li\u003e\n\u003cli\u003eZheng RS, Chen R, Han BF, Wang SM, Li L, Sun KX, et al. [Cancer incidence and mortality in China, 2022]. Zhonghua Zhong Liu Za Zhi. 2024;46(3):221-31.\u003c/li\u003e\n\u003cli\u003eJiang X, Yang L, Chen G, Feng X, Liu Y, Gao Q, et al. Discovery of Kinetin in inhibiting colorectal cancer progression via enhancing PSMB1-mediated RAB34 degradation. Cancer Lett. 2024;584:216600.\u003c/li\u003e\n\u003cli\u003eLeclercq S, de Timary P. Role of the Microbiome and the Gut-Brain Axis in Alcohol Use Disorder: Potential Implication for Treatment Development. Curr Top Behav Neurosci. 2024.\u003c/li\u003e\n\u003cli\u003eGorny RL, Golofit-Szymczak M, Pawlak A, Lawniczek-Walczyk A, Cyprowski M, Stobnicka A, et al. Effectiveness of UV-C radiation in inactivation of microorganisms on materials with different surface structures. Ann Agric Environ Med. 2024;31(2):287-93.\u003c/li\u003e\n\u003cli\u003eWijesekara T, Abeyrathne E, Ahn DU. Effect of Bioactive Peptides on Gut Microbiota and Their Relations to Human Health. Foods. 2024;13(12).\u003c/li\u003e\n\u003cli\u003eBaffy G, Portincasa P. Gut Microbiota and Sinusoidal Vasoregulation in MASLD: A Portal Perspective. Metabolites. 2024;14(6).\u003c/li\u003e\n\u003cli\u003eAbdullah M, Sukartini N, Nursyirwan SA, Pribadi RR, Maulahela H, Utari AP, et al. Gut Microbiota Profiles in Early- and Late-Onset Colorectal Cancer: A Potential Diagnostic Biomarker in the Future. Digestion. 2021;102(6):823-32.\u003c/li\u003e\n\u003cli\u003eHu YZ, Wu CS, Wang J, Han XQ, Si PY, Zhang YA, et al. Antimicrobial Protein LECT2-b Helps Maintain Gut Microbiota Homeostasis via Selectively Targeting Certain Pathogenic Bacteria. J Immunol. 2024;212(1):81-95.\u003c/li\u003e\n\u003cli\u003eLee C, Lee S, Yoo W. Metabolic Interaction Between Host and the Gut Microbiota During High-Fat Diet-Induced Colorectal Cancer. J Microbiol. 2024;62(3):153-65.\u003c/li\u003e\n\u003cli\u003eZhang H, Lan M, Cui G, Zhu W. The Influence of Caerulomycin A on the Intestinal Microbiota in SD Rats. Mar Drugs. 2020;18(5).\u003c/li\u003e\n\u003cli\u003eTajasuwan L, Kettawan A, Rungruang T, Wunjuntuk K, Prombutara P. Role of Dietary Defatted Rice Bran in the Modulation of Gut Microbiota in AOM/DSS-Induced Colitis-Associated Colorectal Cancer Rat Model. Nutrients. 2023;15(6).\u003c/li\u003e\n\u003cli\u003eAvuthu N, Guda C. Meta-Analysis of Altered Gut Microbiota Reveals Microbial and Metabolic Biomarkers for Colorectal Cancer. Microbiol Spectr. 2022;10(4):e0001322.\u003c/li\u003e\n\u003cli\u003eGao Z, Guo B, Gao R, Zhu Q, Qin H. Microbiota disbiosis is associated with colorectal cancer. Front Microbiol. 2015;6:20.\u003c/li\u003e\n\u003cli\u003eHuh JW, Kim MJ, Kim J, Lee HG, Ryoo SB, Ku JL, et al. Enterotypical Prevotella and three novel bacterial biomarkers in preoperative stool predict the clinical outcome of colorectal cancer. Microbiome. 2022;10(1):203.\u003c/li\u003e\n\u003cli\u003eLiu G, Li T, Zhu X, Zhang X, Wang J. An independent evaluation in a CRC patient cohort of microbiome 16S rRNA sequence analysis methods: OTU clustering, DADA2, and Deblur. Front Microbiol. 2023;14:1178744.\u003c/li\u003e\n\u003cli\u003eJin M, Wu J, Shi L, Zhou B, Shang F, Chang X, et al. Gut microbiota distinct between colorectal cancers with deficient and proficient mismatch repair: A study of 230 CRC patients. Front Microbiol. 2022;13:993285.\u003c/li\u003e\n\u003cli\u003eMartin R, Rios-Covian D, Huillet E, Auger S, Khazaal S, Bermudez-Humaran LG, et al. Faecalibacterium: a bacterial genus with promising human health applications. FEMS Microbiol Rev. 2023;47(4).\u003c/li\u003e\n\u003cli\u003eGopalakrishnan V, Spencer CN, Nezi L, Reuben A, Andrews MC, Karpinets TV, et al. Gut microbiome modulates response to anti-PD-1 immunotherapy in melanoma patients. Science. 2018;359(6371):97-103.\u003c/li\u003e\n\u003cli\u003eDai Z, Coker OO, Nakatsu G, Wu WKK, Zhao L, Chen Z, et al. Multi-cohort analysis of colorectal cancer metagenome identified altered bacteria across populations and universal bacterial markers. Microbiome. 2018;6(1):70.\u003c/li\u003e\n\u003cli\u003eWong CC, Yu J. Gut microbiota in colorectal cancer development and therapy. Nat Rev Clin Oncol. 2023;20(7):429-52.\u003c/li\u003e\n\u003cli\u003eShah MS, DeSantis T, Yamal JM, Weir T, Ryan EP, Cope JL, et al. Re-purposing 16S rRNA gene sequence data from within case paired tumor biopsy and tumor-adjacent biopsy or fecal samples to identify microbial markers for colorectal cancer. PLoS One. 2018;13(11):e0207002.\u003c/li\u003e\n\u003cli\u003eOmar Al-Hassi H, Ng O, Brookes M. Tumour-associated and non-tumour-associated microbiota in colorectal cancer. Gut. 2018;67(2):395.\u003c/li\u003e\n\u003cli\u003eMurphy CL, Barrett M, Pellanda P, Killeen S, McCourt M, Andrews E, et al. Mapping the colorectal tumor microbiota. Gut Microbes. 2021;13(1):1-10.\u003c/li\u003e\n\u003cli\u003eQi Z, Zhibo Z, Jing Z, Zhanbo Q, Shugao H, Weili J, et al. Prediction model of poorly differentiated colorectal cancer (CRC) based on gut bacteria. BMC Microbiol. 2022;22(1):312.\u003c/li\u003e\n\u003cli\u003eHuang Z, Huang X, Huang Y, Liang K, Chen L, Zhong C, et al. Identification of KRAS mutation-associated gut microbiota in colorectal cancer and construction of predictive machine learning model. Microbiol Spectr. 2024;12(5):e0272023.\u003c/li\u003e\n\u003cli\u003eKonishi Y, Okumura S, Matsumoto T, Itatani Y, Nishiyama T, Okazaki Y, et al. Development and evaluation of a colorectal cancer screening method using machine learning-based gut microbiota analysis. Cancer Med. 2022;11(16):3194-206.\u003c/li\u003e\n\u003cli\u003eLiu J, Huang X, Chen C, Wang Z, Huang Z, Qin M, et al. Identification of colorectal cancer progression-associated intestinal microbiome and predictive signature construction. J Transl Med. 2023;21(1):373.\u003c/li\u003e\n\u003cli\u003eLuu K, Ye JY, Lagishetty V, Liang F, Hauer M, Sedighian F, et al. Fecal and Tissue Microbiota Are Associated with Tumor T-Cell Infiltration and Mesenteric Lymph Node Involvement in Colorectal Cancer. Nutrients. 2023;15(2).\u003c/li\u003e\n\u003cli\u003eWu Y, Jiao N, Zhu R, Zhang Y, Wu D, Wang AJ, et al. Identification of microbial markers across populations in early detection of colorectal cancer. Nat Commun. 2021;12(1):3063.\u003c/li\u003e\n\u003cli\u003eChen S, Zhang L, Li M, Zhang Y, Sun M, Wang L, et al. Fusobacterium nucleatum reduces METTL3-mediated m(6)A modification and contributes to colorectal cancer metastasis. Nat Commun. 2022;13(1):1248.\u003c/li\u003e\n\u003cli\u003eCui W, Guo M, Liu D, Xiao P, Yang C, Huang H, et al. Gut microbial metabolite facilitates colorectal cancer development via ferroptosis inhibition. Nat Cell Biol. 2024;26(1):124-37.\u003c/li\u003e\n\u003cli\u003eKerdreux M, Edin S, Lowenmark T, Bronnec V, Lofgren-Burstrom A, Zingmark C, et al. Porphyromonas gingivalis in Colorectal Cancer and its Association to Patient Prognosis. J Cancer. 2023;14(9):1479-85.\u003c/li\u003e\n\u003cli\u003eZhong X, Wang Y, Xu J, Cao H, Zhang F, Wang X. Gut microbiota signatures in tissues of the colorectal polyp and normal colorectal mucosa, and faeces. Front Cell Infect Microbiol. 2022;12:1054808.\u003c/li\u003e\n\u003cli\u003eElahi Z, Shariati A, Bostanghadiri N, Dadgar-Zankbar L, Razavi S, Norzaee S, et al. Association of Lactobacillus, Firmicutes, Bifidobacterium, Clostridium, and Enterococcus with colorectal cancer in Iranian patients. Heliyon. 2023;9(12):e22602.\u003c/li\u003e\n\u003cli\u003eAlexander JL, Posma JM, Scott A, Poynter L, Mason SE, Doria ML, et al. Pathobionts in the tumour microbiota predict survival following resection for colorectal cancer. Microbiome. 2023;11(1):100.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"16S rRNA, colorectal cancer, machine-learning algorithm, Gut microbiota","lastPublishedDoi":"10.21203/rs.3.rs-6286866/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6286866/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eObjective: \u003c/strong\u003eThe incidence and mortality of colorectal cancer (CRC) have been increasing, making research into factors related to CRC necessary. This article aims to differentiate characteristics of gut microbiota between CRC patients and healthy individuals, and employs machine-learning algorithms for predicting specific microbial taxa.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e We conducted a multicenter case-control study starting in 2020, used 16S rRNA gene sequencing to analyze the gut microbiota in newly diagnosed CRC patients and healthy individuals. We used Python (version 3.9) to develop predictive models based on machine-learning algorithms.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: Our research indicates a significant abundance of Escherichia-Shigella and Bacteroides in CRC patients, while Blautia and Faecalibacterium notably increased in healthy individuals. Using the Lasso model, we identified eight specific microbial taxa associated with CRC patients and thirteen taxa associated with healthy individuals.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDiscussion\u003c/strong\u003e: The research highlights significant increase of various microbial taxa associated among CRC patients and healthy individuals, and also some microbiota with contentious functionalities. Among the machine-learning algorithms tested, the Random Forest model proved most suitable for predictive modeling in this region.\u003c/p\u003e","manuscriptTitle":"Gut-microbiota in Colorectal Cancer Patients: 16S rRNA Sequencing analysis and Machine-learning Algorithm Prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-07 09:45:46","doi":"10.21203/rs.3.rs-6286866/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"01468d12-1c94-44b7-8f65-91e6a2f2e250","owner":[],"postedDate":"April 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-08-14T06:38:18+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-07 09:45:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6286866","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6286866","identity":"rs-6286866","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}