Machine Learning Meta-analysis of Large Metagenomic Datasets: Tools and Biological Insights
Open accessThe DOI matches its Crossref record; 70 of the 71 references it cites were found in a registry or an index.
Abstract
Shotgun metagenomic analysis of the human associated microbiome provides a rich set of microbial features for prediction and biomarker discovery in the context of human diseases and health conditions. However, the use of such high-resolution microbial features presents new challenges, and validated computational tools for learning tasks are lacking. Moreover, classification rules have scarcely been validated in independent studies, posing questions about the generality and generalization of disease-predictive models across cohorts. In this paper, we comprehensively assess approaches to metagenomics-based prediction tasks and for quantitative assessment of the strength of potential microbiome-phenotype associations. We develop a computational framework for prediction tasks using quantitative microbiome profiles, including species-level relative abundances and presence of strain-specific markers. A comprehensive meta-analysis, with particular emphasis on generalization across cohorts, was performed in a collection of 2424 publicly available metagenomic samples from eight large-scale studies. Cross-validation revealed good disease-prediction capabilities, which were in general improved by feature selection and use of strain-specific markers instead of species-level taxonomic abundance. In cross-study analysis, models transferred between studies were in some cases less accurate than models tested by within-study cross-validation. Interestingly, the addition of healthy (control) samples from other studies to training sets improved disease prediction capabilities. Some microbial species (most notably Streptococcus anginosus) seem to characterize general dysbiotic states of the microbiome rather than connections with a specific disease. Our results in modelling features of the "healthy" microbiome can be considered a first step toward defining general microbial dysbiosis. The software framework, microbiome profiles, and metadata for thousands of samples are publicly available at http://segatalab.cibio.unitn.it/tools/metaml.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures






![Fig 7 Cross-study analysis in multiple gut datasets for (a) T2D discrimination and (b) disease discrimination (independently from the type of disease). For (a), we included all the healthy (controls) and diabetes (cases) samples, whereas samples labelled as other diseases were not considered. For (b), we instead included all the samples where samples with one of the considered diseases were put together in the same "diseases" class. The * denotes cross-validation results (with the margin of error reported in parenthesis). In the other cases, the model was generated on all the datasets other than the dataset considered for testing, a “leave-one-dataset-out” cross-study validation [51]. For the testing datasets with only healthy samples, prediction accuracy was evaluated in terms of overall accuracy (OA). In bold we report the best value for each scenario and feature type (i.e., species abundance or marker presence), and circled are the absolute best value for each scenario.](https://figures.cochat.ai/paper-figures/5bb56b13e2b047c5a1c14f0db710b338/machine-learning-meta-analysis-of-large-metagenomic-datasets-tools-and-figure-7.jpg)
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“We chose to use RF as our classifier, given its, on average, superior performance using microbial feature sets in previous benchmarking studies 31 .”
From A gut microbial signature for combination immune checkpoint blockade across cancer types (2024) · cites this paper as [31]
In the citing paper
“As in previous studies (Pasolli et al., 2016; Thomas et al., 2019), RFs using functional features performed similarly (0.69 cross validation and 0.71 LODO ROC AUC on pathways relative abundance), indicating a tight link between strain-specific taxonomy and gene carriage in this setting.”
From Integrating taxonomic, functional, and strain-level profiling of diverse microbial communities with bioBakery 3 (2021) · cites this paper as Pasolli et al., 2016
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Breiman, L. (2001) · Machine Learning · doi:10.1023/a:1010933404324 · Cited by 133,538
Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016 · cited 1×
“The classifiers were implemented using the scikit-learn python package [61].”
MethodCortes, C., & Vapnik, V. (1995) · Machine Learning · doi:10.1007/bf00994018 · Cited by 59,421
Regression shrinkage and selection via the lasso
Tibshirani, R. (1996) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1996.tb02080.x · Cited by 51,482 · cited 13×
“Lasso is based on a multiple logistic regression trained with L1-norm penalized likelihood, while both L1 and L2 norms are penalized in ENet.”
BackgroundMethodInfluentialNeural Networks: A Comprehensive Foundation
Haykin, S. (1998) · Cited by 30,070
Wu, Y. (1999) · Technometrics · doi:10.1080/00401706.1999.10485951 · Cited by 27,221
The Elements of Statistical Learning
Hastie, T., Friedman, J., & Tibshirani, R. (2001) · Springer Series in Statistics · doi:10.1007/978-0-387-21606-5 · Cited by 24,162
Regularization and Variable Selection Via the Elastic Net
Zou, H., & Hastie, T. (2005) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2005.00503.x · Cited by 20,957
Classification and Regression Trees.
Ryzin, J. V., Breiman, L., Friedman, J. H., Olshen, R. A., & Stone, C. J. (1986) · Journal of the American Statistical Association · doi:10.2307/2288003 · Cited by 20,825
Metagenomic biomarker discovery and explanation
Segata, N., Izard, J., Waldron, L., Gevers, D., Miropolsky, L., Garrett, W. S., & Huttenhower, C. (2011) · Genome Biology · doi:10.1186/gb-2011-12-6-r60 · Cited by 16,192 · cited 1×
“The approach proposed here (RF with embedded feature selection) focuses on the set of features with the most discriminating power rather than on strictly statistical assessments [46] or statistical assessment coupled with effect size [47].”
MethodSelf-citation
Research works printed but not linked: 1
- J Qin, 2012, A metagenome-wide association study of gut microbiota in type 2 diabetes, Nature, 490, 55, 10.1038/nature11450 · DOI · Paper page
Cited by
Items 301–400 of 666
Chang, C.-C., Liu, T.-C., Lu, C.-J., Chiu, H., & Lin, W. (2023) · Frontiers in Microbiology · doi:10.3389/fmicb.2023.1227300 · Cited by 17
“Traditional ML models, including linear regression with Lasso and elastic nets, have been demonstrated to have higher performance in analyzing gut microbiome data and predicting dysbiosis (Pasolli et al., 2016; Lee and Rho, 2022).”
MethodA primer and discussion on DNA-based microbiome data and related bioinformatics analyses
Douglas, G. M., & Langille, M. G. I. (2021) · Peer Community Journal · doi:10.24072/pcjournal.2 · Cited by 17
Valles-Colomer, M., Manghi, P., Cumbo, F., Masetti, G., Armanini, F., Asnicar, F., Blanco-Miguez, A., Pinto, F., Punčochář, M., Garaventa, A., et al. (2024) · eBioMedicine · doi:10.1016/j.ebiom.2023.104917 · Cited by 14
“To classify the individuals in different groups based on their gut microbiome composition, two random forest classi fi ers were built with MetAML 33 (which implements the scikit-learn python library Random Forest (RF) 34 ), similarly to the method used in. 11,12,35 The fi rst random forest model was built to classify sa…”
MethodInfluentialSelf-citationColorectal microbiota after removal of colorectal cancer
Cronin, P., Murphy, C. L., Barrett, M., Ghosh, T. S., Pellanda, P., O’Connor, E. M., Zulquernain, S. A., Kileen, S., McCourt, M., Andrews, E. J., et al. (2022) · NAR Cancer · doi:10.1093/narcan/zcac011 · Cited by 16
Lima, H. S., de Oliveira, G. F. V., Ferreira, R. D. S., de Castro, A. G., Silva, L. C. F., Ferreira, L. D. S., Oliveira, D. A. D. S., da Silva, L. F., Kasuya, M. C. M., de Paula, S. O., et al. (2024) · Journal of Environmental Management · doi:10.1016/j.jenvman.2024.120559 · Cited by 15
Dhungel, E., Mreyoud, Y., Gwak, H., Rajeh, A. A., Rho, M., & Ahn, T.-H. (2021) · BMC Bioinformatics · doi:10.1186/s12859-020-03933-4 · Cited by 15
Arukha, A. P., Freguia, C. F., Mishra, M., Jha, J. K., Kariyawasam, S., Fanger, N. A., Zimmermann, E. M., Fanger, G. R., & Sahay, B. (2021) · Biomedicines · doi:10.3390/biomedicines9091098 · Cited by 17
Mining Archive.org’s Twitter Stream Grab for Pharmacovigilance Research Gold
Tekumalla, R., Asl, J. R., & Banda, J. M. (2020) · Proceedings of the International AAAI Conference on Web and Social Media · doi:10.1609/icwsm.v14i1.7357 · Cited by 14
Mathematical properties of optimal fluxes in cellular reaction networks at balanced growth
Dourado, H., Liebermeister, W., Ebenhöh, O., & Lercher, M. J. (2023) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1011156 · Cited by 14
Monshizadeh, M., & Ye, Y. (2024) · Gut Microbes · doi:10.1080/19490976.2024.2302076 · Cited by 13
Bazzani, D., Heidrich, V., Manghi, P., Blanco‐Míguez, A., Asnicar, F., Armanini, F., Cavaliere, S., Bertelle, A., Dell'Acqua, F., Dellasega, E., et al. (2024) · npj Biofilms and Microbiomes · doi:10.1038/s41522-024-00482-z · Cited by 17
Self-citationWang, B., & Luan, Y. (2024) · Frontiers in Genetics · doi:10.3389/fgene.2024.1369628 · Cited by 14
“Compared with SVM and LASSO, it has been shown to outperform them when applied to microbiome data (Pasolli et al., 2016).”
MethodMurovec, B., Deutsch, L., & Stres, B. (2024) · Frontiers in Microbiology · doi:10.3389/fmicb.2024.1426407 · Cited by 14
“The human gut microbiome, which encompasses the microbial communities in the intestinal tract, is becoming increasingly important because of its role in human disease (Pasolli et al., 2016).”
BackgroundKim, N., Kim, C. Y., Ma, J., Yang, S., Park, D. J., Ha, S., Belenky, P. A., & Lee, I. (2024) · Gut Microbes · doi:10.1080/19490976.2024.2393791 · Cited by 15
“We adopted the random forest algorithm as implemented in metaML, 58 following the parameters outlined by Thomas et al. 59 The script for this analysis was obtained from the MetAML GitHub repository (https://github.com/ SegataLab/metaml).”
MethodChen, C., & Tyler, A. D. (2020) · Biology Direct · doi:10.1186/s13062-020-00287-y · Cited by 13
Human Gut Microbiota in Heart Failure: Trying to Unmask an Emerging Organ
PARASKEVAIDIS, I. A., Xanthopoulos, A. V., Tsougos, E., & Triposkiadis, F. K. (2023) · Biomedicines · doi:10.3390/biomedicines11092574 · Cited by 14
Rahman, M. A., & Rangwala, H. (2020) · Bioinformatics · doi:10.1093/bioinformatics/btaa477 · Cited by 16
Vänni, P., Tejesvi, M. V., Ainonen, S., Renko, M., Korpela, K., Salo, J., Paalanne, N., & Tapiainen, T. (2021) · Scientific Reports · doi:10.1038/s41598-021-97007-x · Cited by 16
Li, L., Yan, T., Yan, R., Zeng, X., Du, Q., Wu, Z. F., & Pan, D. (2025) · Food & Function · doi:10.1039/d5fo01850a · Cited by 15
Soluble MAdCAM-1 as a biomarker in metastatic renal cell carcinoma
Silva, C. A. C., Machaalani, M., Saliby, R. M., Zhong, C., Xie, W., Pasolli, E., Piccinno, G., Dalban, C., Fidelle, M., Meurisse, A., et al. (2026) · Nature Medicine · doi:10.1038/s41591-025-04067-x · Cited by 17
“Large MGS datasets have enabled the development of microbiota-based biomarkers based on machine learning 23 or custom scoring based on quantitative polymerase chain reaction 24 .”
Self-citationKhachatryan, L., Xiang, Y., Ivanov, A. B., Glaab, E., Graham, G. T., Granata, I., Giordano, M., Maddalena, L., Piccirillo, M., Manipur, I., et al. (2023) · Scientific Reports · doi:10.1038/s41598-023-33050-0 · Cited by 13
Translating the Microbiome: What’s the Target?
O’Toole, P. W., Ghosh, T. S., Goswami, S., Manghi, P., Segata, N., & Shanahan, F. (2023) · Gastroenterology · doi:10.1053/j.gastro.2023.04.008 · Cited by 17
Self-citationWang, Q., Fan, X., Wu, S., & Su, X. (2024) · Bioinformatics Advances · doi:10.1093/bioadv/vbae013 · Cited by 14
“Consequently, machine learning (ML) techniques have been widely employed to uncover the connections between microbial dynamics and human health status (Forslund et al. 2015, Pasolli et al. 2016, Qu et al. 2019, Fenglong et al. 2021).”
PM2RA: A Framework for Detecting and Quantifying Relationship Alterations in Microbial Community
Liu, Z., Mi, K., Xu, Z. Z., Zhang, Q., & Liu, X. (2021) · Genomics Proteomics & Bioinformatics · doi:10.1016/j.gpb.2020.07.005 · Cited by 13
“Two diabetes datasets were downloaded from the published datasets [32].”
Meta-Signer: Metagenomic Signature Identifier based onrank aggregation of features
Reiman, D. J., Metwally, A. A., Sun, J., & Dai, Y. D. (2021) · F1000Research · doi:10.12688/f1000research.27384.1 · Cited by 12
Olekhnovich, E. I., Ivanov, A. B., Игоревич, У. В., & Ilina, E. N. (2021) · mSystems · doi:10.1128/msystems.00811-21 · Cited by 13
“For additional computational experiments, 139 metagenomic stool samples from the HMP 2012 data set (41) were used.”
MethodMegaD: Deep Learning for Rapid and Accurate Disease Status Prediction of Metagenomic Samples
Mreyoud, Y., Song, M., Lim, J., & Ahn, T.-H. (2022) · Life · doi:10.3390/life12050669 · Cited by 15
“MetAML takes advantage of several machine learning models to link taxonomic profiles to specific phenotypes [24].”
BackgroundSupervised Machine Learning Enables Geospatial Microbial Provenance
Bhattacharya, C., Tierney, B. T., Ryon, K. A., Bhattacharyya, M., Hastings, J. J. A., Basu, S., Bhattacharya, B., Bagchi, D., Mukherjee, S., Wang, L., et al. (2022) · Genes · doi:10.3390/genes13101914 · Cited by 13
Developing a New Phylogeny-Driven Random Forest Model for Functional Metagenomics
Wassan, J. T., Wang, H., & Zheng, H. (2023) · IEEE Transactions on NanoBioscience · doi:10.1109/tnb.2023.3283462 · Cited by 13
“Using Random Forest(RF) classifier for classifying genotypes into phenotypes has been popularized [4], [5], [6], [7], [8], [9], [10], [11].”
MethodRegMIL: Phenotype Classification from Metagenomic Data
Rahman, M. A., & Rangwala, H. (2018) · ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB) · doi:10.1145/3233547.3233585 · Cited by 18
“We compared our proposed method RegMIL with non-MIL approaches such as method [28] proposed by Qin et al. and MetAML [24] which were briefly discussed in Section 2.”
MethodResultInfluentialMétris, A., Barrett, P., Price, L. J., Klamert, S., & Fernandez‐Piquer, J. (2021) · Microbial Risk Analysis · doi:10.1016/j.mran.2021.100188 · Cited by 14
Machine and Deep Learning in Molecular and Genetic Aspects of Sleep Research
Elgart, M., Redline, S. S., & Sofer, T. (2021) · Neurotherapeutics · doi:10.1007/s13311-021-01014-9 · Cited by 12
Evaluating supervised and unsupervised background noise correction in human gut microbiome data
Briscoe, L., Balliu, B., Sankararaman, S., Halperin, E., & Garud, N. R. (2022) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1009838 · Cited by 12
“The pooling of datasets is a major contributor of confounding yet combining datasets is an increasingly common [1,25–28] and powerful means to validate associations [8,29] in a discovery dataset with held out datasets [1,30,31].”
BackgroundMethodKGA: integrating KPCA and GAN for microbial data augmentation
Wen, L.-Y., Zhang, X., Li, Q., & Min, F. (2022) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-022-01707-3 · Cited by 12
“Random forest (RF) is used for building prediction models for colorectal cancer, liver cirrhosis, and other diseases [1, 33].”
MethodSkorentseva, K. V., Bolshakov, F. V., Saidova, A. A., & Lavrov, A. (2023) · Cell and Tissue Research · doi:10.1007/s00441-023-03810-5 · Cited by 15
Reiman, D. J., & Dai, Y. D. (2020) · International Conference on Deep Learning Theory and Applications (DeLTA) · doi:10.5220/0009892601030110 · Cited by 11
Towards end-to-end disease prediction from raw metagenomic dataPreprint
Queyrel, M., Prifti, E., Templier, A., & Zucker, J. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.10.29.360297 · Cited by 12
“The datasets are composed by only MetaML [43] and DeepMicro [46] are the reference methods.”
BackgroundMethodInfluentialVänni, P., Tejesvi, M. V., Paalanne, N., Aagaard, K. M., Ackermann, G. L., Camargo, C. A., Eggesbø, M. Å., Hasegawa, K., Hoen, A. G., Karagas, M. R., et al. (2023) · mSystems · doi:10.1128/msystems.00364-23 · Cited by 11
“…number of studies combining or comparing gut microbiome data from several available prospective cohort studies in neonates, infants, or children using the ML approach, such as random forest (17–20), support vector machine (17, 18), elastic net (17, 18), and gradient-boosted machine algorithms (18).”
BackgroundMethodResultInfluentialArtificial intelligence: the human response to approach the complexity of big data in biology
Melandri, G., Radohery, G. F. R., Beaumont, C., de Cripan, S. M., Muller, C., Piras, L., Pereira, M. A., Salvador, A. F., Domingo-Almenara, X., Bolger, M., et al. (2025) · GigaScience · doi:10.1093/gigascience/giaf057 · Cited by 11
“AI has been in-cr easingl y a pplied to help inter pr et the v ariations found in mi-cr obial comm unities, particularl y in the human micr obiome and its relationship to health and disease [ 69–71 ].”
CRCpred: An AI-ML tool for colorectal cancer prediction using gut microbiome
Pateriya, D., Malwe, A. S., & Sharma, V. K. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.110592 · Cited by 13
Embracing the mantra of modellers and synthesizing omics, experiments and models
Dick, G. J. (2016) · Environmental Microbiology Reports · doi:10.1111/1758-2229.12491 · Cited by 10
“…and distributions in metagenomic datasets serve as markers of microbially mediated processes – whether we’re talking about cryptic biogeochemical cycles (Canfield et al., 2010), transformations of toxins and pollutants (Podar et al., 2015) or markers of human diseases (Pasolli et al., 2016).”
BackgroundApplication of Deep Learning in Microbiome
Zhu, Q., Huo, B., Sun, H. L., Li, B., & Jiang, X. (2020) · Journal of Artificial Intelligence for Medical Sciences · doi:10.2991/jaims.d.201028.001 · Cited by 11
“Pasolli et al. [21] developed a machine learning analysis framework to assess the association between microbiome and phenotypes.”
BackgroundA comparison of dimensionality reduction methods for large biological data
Babjac, A., Royalty, T. M., Steen, A. D., & Emrich, S. (2022) · ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB) · doi:10.1145/3535508.3545536 · Cited by 10
“Specifically, we use an autoencoders’ latent space representation in combination with classification algorithms, similar to how MetAML [21] pairs either elastic net or LASSO-based feature selection with classification.”
BackgroundMethodInfluentialMetagenomic Geolocation Prediction Using an Adaptive Ensemble Classifier
Anyaso-Samuel, S., Sachdeva, A., Guha, S., & Datta, S. (2021) · Frontiers in Genetics · doi:10.3389/fgene.2021.642282 · Cited by 13
A Fast Machine Learning Workflow for Rapid Phenotype Prediction from Whole Shotgun Metagenomes
Carrieri, A. P., Rowe, W. P. M., Winn, M., & Pyzer‐Knapp, E. O. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33019434 · Cited by 11
“A 2016 machine learning framework (Pasolli et al. 2016) uses quantitative microbiome profiles, including species-level relative abundances and presence/absence of species-and strain-specific markers, as features for ML models.”
MethodSakowski, E. G., Uritskiy, G., Cooper, R., Gomes, M. L., McLaren, M. R., Meisel, J. S., Mickol, R. L., Mintz, C. D., Mongodin, E. F., Pop, M., et al. (2019) · mSystems · doi:10.1128/msystems.00392-19 · Cited by 13
The comparable strategic flexibility of model-free and model-based learningPreprint
Filipowicz, A. L. S., Levine, J., Piasini, E., Tavoni, G., Kable, J. W., & Gold, J. I. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2019.12.28.879965 · Cited by 11
Hacılar, H., Nalbantoğlu, Ö. U., Aran, O., & Bakır-Güngör, B. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.03019 · Cited by 10
“In this respect, feature selection methodologies such as mRMR [15], Lasso [7], Elastic Net [13] and iterative sure select algorithm [15] have been extensively applied to reduce the number of taxa (species).”
MethodInfluentialBioinformatic approaches to blood and tissue microbiome analyses: challenges and perspectives
Sirasani, J. P., Gardner, C., Jung, G., Lee, H., & Ahn, T.-H. (2025) · Briefings in Bioinformatics · doi:10.1093/bib/bbaf176 · Cited by 11
“MetAML leverages multiple ML models to associate taxonomic profiles with specific phenotypes [104].”
Youngblut, N. D., de la Cuesta‐Zuluaga, J., & Ley, R. E. (2022) · Environmental Microbiology · doi:10.1111/1462-2920.15910 · Cited by 12
Imangaliyev, S., Schlötterer, J., Meyer, F., & Seifert, C. (2022) · Diagnostics · doi:10.3390/diagnostics12102514 · Cited by 10
Cao, C., Jin, X., Ding, Q., Zhu, J., Yang, D., & Fan, B. (2023) · Urolithiasis · doi:10.1007/s00240-023-01467-x · Cited by 11
“With the development of artificial intelligence, an increasing number of machine learning models have been employed in disease prediction research [61, 62].”
MethodGao, W., Wang, X., Shi, Y., Wu, G., Zhou, M., & Lin, X. (2024) · Genes and Immunity · doi:10.1038/s41435-024-00306-2 · Cited by 9
“Other unspeci fi ed statistical tests were performed by Wilcoxon test.”
RetractedAn Ensemble Feature Selection Method Based on Deep Forest for Microbiome-Wide Association Studies
Zhu, Q., Pan, M., Liu, L., Li, B., He, T., Jiang, X., & Hu, X. T. (2018) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm.2018.8621461 · Cited by 10
“We use three different data types from MetAML package [16]: cirrhosis, type 2 diabetes (T2D) and obesity.”
MethodNguyen, T. H. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-35653-8_20 · Cited by 9
Shi, K., Zhang, L., Yu, J., Chen, Z., Lai, S., Zhao, X., Li, W., Luo, Q., Lin, W., Feng, J., et al. (2021) · Clinical and Translational Medicine · doi:10.1002/ctm2.314 · Cited by 9
Microbiome and colorectal cancer: A review of the past, present, and future
Johns, M. S., & Petrelli, N. J. (2021) · Surgical Oncology · doi:10.1016/j.suronc.2021.101560 · Cited by 10
Decoding gut microbiota by imaging analysis of fecal samples
Furusawa, C., Tanabe, K., Ishii, C., Kagata, N., Tomita, M., & Fukuda, S. (2021) · iScience · doi:10.1016/j.isci.2021.103481 · Cited by 8
“In such analyses, machine learning algorithms, including random forest and support vector machine algorithms, are often used to classify and extract low-dimensional features (Duvallet et al., 2017; Pasolli et al., 2016; Roguet et al., 2018).”
BackgroundNaseri, M., Palizban, F., Yadegar, A., Khodarahmi, M., Aghdaei, H. A., Houri, H., & Zahiri, J. (2022) · Gut Pathogens · doi:10.1186/s13099-022-00482-4 · Cited by 9
“Nowadays, the metagenomics analysis of the human-associated microbiome provides a rich set of microbial features for prediction and biomarker discovery in the context of human diseases and health conditions [4].”
BackgroundMB-SupCon: Microbiome-based Predictive Models via Supervised Contrastive Learning
Yang, S., Wang, S., Wang, Y., Rong, R., Kim, J., Li, B., Koh, A. Y., Xiao, G., Li, Q., Liu, D. J., et al. (2022) · Journal of Molecular Biology · doi:10.1016/j.jmb.2022.167693 · Cited by 9
[Therapeutic Relevance of Human Microbiota and Lung Cancer].
Zhou, H., Suo, J., & Zhu, J. (2019) · PubMed Central · doi:10.3779/j.issn.1009-3419.2019.07.09 · Cited by 11
Data-science ready, multisite, human diffusion MRI white-matter-tract statistics
Lerma‐Usabiaga, G., Mukherjee, P., Perry, L. M., & Wandell, B. A. (2020) · Scientific Data · doi:10.1038/s41597-020-00760-3 · Cited by 10
The role of the gut microbiome in regulating the response to immune checkpoint inhibitor therapy
Dean, N. J., d’Arienzo, P. D., Ibraheim, H., Lee, K. A., Olsson-Brown, A. C., Pinato, D. J., & Powell, N. (2024) · Best Practice & Research Clinical Gastroenterology · doi:10.1016/j.bpg.2024.101944 · Cited by 9
Postmortem microbiome dynamics: Review of forensic microbial clock
Kalanjali, Y., & Isukapatla, A. R. (2025) · Journal of Forensic and Legal Medicine · doi:10.1016/j.jflm.2025.103024 · Cited by 14
Omics-based decoding of molecular and metabolic crosstalk in the skin barrier ecosystem
Elettrico, L., Piacenti, G., Levron, C. L., Ansai, O., Croce, A., Duval, C., Proserpio, V., & Donati, G. (2026) · Cell Death and Differentiation · doi:10.1038/s41418-025-01648-8 · Cited by 8
“TensorFlow [132] and PyTorch [133], two open-source deep learning frameworks, have been applied to build models predicting associations between microbial compositions and epithelial health [134, 135].”
Kwak, M., Park, B., Choi, H., Hong, W., Mun, D., Son, S.-H., Choi, Y.-S., Pathiraja, D., Eor, J. Y., Ryu, S., et al. (2026) · Pharmacological Research · doi:10.1016/j.phrs.2026.108184 · Cited by 10
Agostinetto, G., Anna, S., Antonia, B., Dario, P., & Casiraghi, M. (2022) · Frontiers in Bioinformatics · doi:10.3389/fbinf.2021.794547 · Cited by 10
“Spanning from classification and signature extraction to interaction and trait associations (Pasolli et al., 2016; Qu et al., 2019), data mining strategies can identify hidden patterns that may help to predict biological functions (Noor et al.”
BackgroundPatterson, A., Elbasir, A., Tian, B., & Auslander, N. (2023) · Cancers · doi:10.3390/cancers15071958 · Cited by 9
Yang, Z., Zhu, J., Lu, W., Tian, F., Zhang, H., & Chen, W. (2024) · Food Bioscience · doi:10.1016/j.fbio.2024.104541 · Cited by 10
Bašić-Čičak, D., Telalović, J. H., & Pašić, L. (2024) · Diagnostics · doi:10.3390/diagnostics14222536 · Cited by 8
Elhag, M. R. A., Abu-Median, A.-B., Boit, M. D., Haris, P. I., Madkour, M. I., Alkawamleh, D. H., & Faris, M. E. (2025) · Clinical Nutrition ESPEN · doi:10.1016/j.clnesp.2025.06.051 · Cited by 9
Coppola, S., Paparo, L., Chiariotti, L., Ercolini, D., Nocerino, R., Severina, A. F. D. G. D. S., Carucci, L., Filippis, F. D., Agangi, A., Napolitano, M., et al. (2022) · Frontiers in Nutrition · doi:10.3389/fnut.2022.951223 · Cited by 9
“Machine learning-based classification analysis will be done using the MetAML package (30) and by considering random forests (RFs) as back-end classifier for all the results.”
MethodLiang, C., Wagstaff, J., Aharony, N., Schmit, V. L., & Manheim, D. (2023) · Health Security · doi:10.1089/hs.2022.0029 · Cited by 8
Hajihosseini, M., Amini, P., Saidi‐Mehrabad, A., & Dinu, I. A. (2023) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2023.02.027 · Cited by 9
“Machine learning methods are still developing in microbiome literature as they are well-known for feature selection and high-dimension data analyses [83,84].”
BackgroundHao, Y. (2025) · Modern Rheumatology · doi:10.1093/mr/roaf081 · Cited by 10
Hashem, H. R., Yehia, T., Azab, M., Abdellah, A., Amin, I. A., Salah, M., & Ramadan, M. (2025) · Gut Pathogens · doi:10.1186/s13099-025-00761-w · Cited by 8
“The identification of potential microbial classifiers associated with HCC was achieved using machine learning methods validated in microbiome research [46].”
MicroHDF: predicting host phenotypes with metagenomic data using a deep forest-based framework
Shi, K., Liu, Q., Ji, Q., He, Q., & Zhao, X.-M. (2024) · Briefings in Bioinformatics · doi:10.1093/bib/bbae530 · Cited by 7
“[17] proposed a comprehensive meta-analysis framework that applied cross-validation, cross-study validation, and cross-disease validation to ML methods.”
MethodModeling microbiome-trait associations with taxonomy-adaptive neural networks
Jiang, Y., Aton, M., Zhu, Q., & Lu, Y. Y. (2025) · Microbiome · doi:10.1186/s40168-025-02080-3 · Cited by 13
“Here, a microbial sample is typically characterized by its taxonomic profile, which includes the abundance of microbial taxa at certain taxonomic levels [9], such as species, genus, family, and so on.”
Ravichandar, J. D., Rutherford, E. M., Chow, C. T., Han, A., Yamamoto, M. L., Narayan, N., Kaplan, G. G., Beck, P. L., Claesson, M. J., Dabbagh, K., et al. (2022) · Frontiers in Microbiology · doi:10.3389/fmicb.2022.961020 · Cited by 7
“Appreciation of strain-specific markers in single-technology meta-analysis workflows has been noted (Pasolli et al., 2016), and given that biological activity can be strain-specific, knowledge of specific strains is fundamental to deciphering their functional role in disease (De Filippis et al.”
BackgroundPractical sampling of constraint-based models: Optimized thinning boosts CHRR performance
Jadebeck, J. F., Wiechert, W., & Nöh, K. (2023) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1011378 · Cited by 11
Host Phenotype Prediction from Differentially Abundant Microbes Using RoDEO
Carrieri, A. P., Haiminen, N., & Parida, L. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-67834-4_3 · Cited by 7
Diagnosis Approaches for Colorectal Cancer Using Manifold Learning and Deep Learning
Nguyễn, H. T., & Thai-Nghe, N. (2020) · SN Computer Science · doi:10.1007/s42979-020-00297-7 · Cited by 7
Making Sense of a Scent-Sensing Metaphor for Microbes and Environmental Predictions
Bowman, J. S. (2021) · mSystems · doi:10.1128/msystems.00993-21 · Cited by 7
Lung cancer and the Gut-microbiota-lung Axis: emerging evidence and potential clinical implications
Liu, L., Li, Y., Zhang, H., Li, H., Shang, T., & Liu, L. (2025) · Frontiers in Medicine · doi:10.3389/fmed.2025.1655780 · Cited by 8
Food systems microbiome‐related educational needs
Olmo, R., Wetzels, S. U., Berg, G., Cocolin, L., Hartmann, M., Hugas, M., Kostić, T., Rattei, T., Ruthsatz, M., Rybakova, D., et al. (2023) · Microbial Biotechnology · doi:10.1111/1751-7915.14263 · Cited by 9
“This led to studies with higher statistical power, and consequently to the transition of operational taxonomic unit tables and functional profiles from end- goal deliverables into starting material for downstream analyses such as machine learning (ML) applications (Pasolli et al., 2016).”
BackgroundLiu, D., Zhou, H., Qu, Y., Zhang, H., & Xu, Y. (2025) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbbio.2025.3546729 · Cited by 10
“Pasolli et al. [17] comprehensively evaluated the prediction tasks based on shotgun metagenomics and the methods for microbial phenotype association evaluation, primarily employing support vector machines and random forest models for disease prediction, with the help of Lasso and elastic net (ENet)…”
Nguyễn, H. T., Phan, T. T., Dao, C.-T., Ta, P. V. D., Nguyen, C. N. T., Pham, N. H., & Huynh, H. X. (2021) · Journal of Engineering and Technological Sciences · doi:10.5614/j.eng.technol.sci.2021.53.1.9 · Cited by 6
“In other studies, researchers used algorithms such as mRMR (Min Redundancy Max Relevance) [14], Lasso (Least Absolute Shrinkage and Selection Operator) [15], Elastic Net [16].”
MethodLi, B., Zhong, D., Jiang, X., & He, T. (2021) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm52615.2021.9669509 · Cited by 7
MetaAnalyst: a user-friendly tool for metagenomic biomarker detection and phenotype classification
AlShawaqfeh, M. K., Rababah, S., Hayajneh, A., Gharaibeh, A. M., & Serpedin, E. (2022) · BMC Medical Research Methodology · doi:10.1186/s12874-022-01812-5 · Cited by 6
“…algorithms and computational tools have been proposed for biomarker detection such as LEfSe [9], RPCA [10], RegLRSD [11], IMG/M [12], MeAtML [13], Fizzy [14], Boruta [15], ENNB [16], MetagenomeSeq [17], MicrobiomeDDA [18], Shotgun-FunctionalizeR [19], MetaStats [20], Raida [21], FAN-TOM…”
MethodCramer, T. (2023) · Seminars in Immunology · doi:10.1016/j.smim.2023.101736 · Cited by 6
Identifying Heat-Resilient Corals Using Machine Learning and Microbiome
Yong, H., & Oudah, M. (2023) · Lecture notes in networks and systems · doi:10.1007/978-3-031-38079-2_6 · Cited by 6
Popa, O., Oldenburg, E., & Ebenhöh, O. (2020) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.2019.0448 · Cited by 6
“Today several methods already exist for particular sequence analysis [152 – 154].”
MethodNguyễn, H. T., Bao, T., Hoang, H., Phuoc, T., & Nghi, C. (2020) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2020.0110607 · Cited by 6
“The the jet colormap exhibits a quite good and reaches the highest average performance on four datasets while gray scale works well on OBE and WT2 and the rainbow achieves the highest performance on COL while the custom colormap gives exceptional results on CIR with the performances of 0.926.”
BackgroundMethodInfluentialMixed Effect Dirichlet-Tree Multinomial for Longitudinal Microbiome Data and Weight PredictionPreprint
Tang, Y., & Nicolae, D. L. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.06380 · Cited by 7
“Directly applying existing methods such as support vector machine and random forest (Pasolli et al., 2016) can lead to loss of prediction accuracy or results hard to interpret.”
MethodZhang, J., Wang, X., Wang, D., Zheng, Z., Wang, H., & La, M. (2026) · mSystems · doi:10.1128/msystems.00800-25 · Cited by 6
“Although ML has been preliminarily applied to identify specific microbial taxa and has yielded valuable insights, it still faces several challenges (44, 112).”
Cui, Z., Wu, Y. X., Zhang, Q., Wang, S., He, Y., & Huang, D.-S. (2023) · Frontiers in Microbiology · doi:10.3389/fmicb.2023.1238199 · Cited by 6
“For SVM and RF, we used the same grid search to set the hyperparameters, referring to MetAML (Pasolli et al., 2016).”
MethodInterpretable metric learning in comparative metagenomics: The adaptive Haar-like distance
Gorman, E. D., & Lladser, M. E. (2024) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1011543 · Cited by 8
“RFs usually exhibit superior or comparative performance to other state-of-the-art methods in microbiome host-trait prediction [16], and numerous studies have documented their effectiveness in metagenomics classification tasks [17–22].”
BackgroundBekhet, S., Alsheref, F. K., AbdelAziz, A. M., & Alshazly, H. A. (2025) · Multimedia Tools and Applications · doi:10.1007/s11042-025-20933-9 · Cited by 5
“This is because fea - tures in microbial data tend to be complex and far exceed the quantity of accessible samples [23], which can lead to model over-fitting that provides false evaluations.”
An, T.-F., Zhang, S., Li, J., Wang, H., Chen, L., Shi, Y., Wang, J., Han, S., Wang, R., Wang, L., et al. (2025) · Frontiers in Microbiology · doi:10.3389/fmicb.2025.1660775 · Cited by 7
DeepGum: Deep feature transfer for gut microbiome analysis using bottleneck models
Çiftcioğlu, U. G. E., & Nalbantoğlu, Ö. U. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2024.105984 · Cited by 6
Metadata
- Authors
- Pasolli, E., Truong, D. T., Malik, F., Waldron, L., & Segata, N.
- Affiliations
- University of Trento (IT)
- City University of New York (US)
- Corresponding author
- Segata, Nicola
- Venue
- PLOS Computational Biology
- Publisher
- Public Library of Science (PLoS)
- Published
- 2016-07-11
- Volume
- 12
- Issue
- 7
- Pages
- e1004977
- ISSN
- 1553-734X (ISSN-L)
- 1553-7358
- PMC
- PMC4939962
- OpenAlex
- W2473355215
- Cited by
- 659
- References
- 71
- Citations by year
- 2026: 81 · 2025: 66 · 2024: 83 · 2023: 64 · 2022: 84 · 2021: 93 · 2020: 80 · 2019: 44 · 2018: 38 · 2017: 24 · 2016: 1
- FWCI
- 3.23
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- PLoS Computational Biology (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Iris (University of Trento) (submittedVersion, cc-by)
- RePEc: Research Papers in Economics (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Figshare (submittedVersion, cc-by)
- Funders
- Ministero dell’Istruzione, dell’Università e della Ricerca (RBFR13EWWI)
- Seventh Framework Programme (BE) (PCIG13-618833)
- Fondazione Caritro (Rif.Int.2013.0239)
- Terme di Comano
- National Science Foundation (CNS-958379, CNS-0855217, ACI-1126113)
- City University of New York High Performance Computing Center
- National Institute of Allergy and Infectious Diseases (1R21AI121784-01)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gut microbiota and health
- Health Sciences › Dentistry › Periodontics › Oral microbiology and periodontitis research
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
- Keywords
- metagenomics, human microbiome, disease prediction, biomarker discovery, strain-specific markers, dysbiosis, Streptococcus anginosus, feature selection, cross-validation
- MeSH terms
- Gastrointestinal Microbiome, Machine Learning, Humans, Obesity, Software, Colorectal Neoplasms, Inflammatory Bowel Diseases, Computational Biology, Metagenome, Metagenomics
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.