Microbiome meta-analysis and cross-disease comparison enabled by the SIAMCAT machine learning toolbox
Year: 2021
Type: article
Abstract: The human microbiome is increasingly mined for diagnostic and therapeutic biomarkers using machine learning (ML). However, metagenomics-specific software is scarce, and overoptimistic evaluation and limited cross-study generalization are prevailing issues. To address these, we developed SIAMCAT, a versatile R toolbox for ML-based comparative metagenomics. We demonstrate its capabilities in a meta-analysis of fecal metagenomic studies (10,803 samples). When naively transferred across studies, ML models lost accuracy and disease specificity, which could however be resolved by a novel training set augmentation strategy. This reveals some biomarkers to be disease-specific, with others shared across multiple conditions. SIAMCAT is freely available from siamcat.embl.de .
Source: Genome biology
Institutions: European Molecular Biology Laboratory, European Molecular Biology Laboratory, European Molecular Biology Laboratory, European Molecular Biology Laboratory, European Molecular Biology Laboratory, Clinical Microbiomics (Denmark), Max Delbrück Center, Charité – Universitätsmedizin Berlin, University of Trento, SIB Swiss Institute of Bioinformatics, ETH Zurich, Institute of Microbiology, University of Würzburg
Language: English
Topic: Gut microbiota and health
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $4,890
Awards: G2813928278, G306507530, G524245738, G5834875682, CompLS - Runde 1 - Verbundprojekt: LAMarCK - Charakterisierung von Krankheiten mittels longitudinaler Multi-Omics Analyse - Teilprojekt A, SFB 1371: Microbiome Signatures Funktionelle Relevanz des Mikrobioms im Verdauungstrakt, de.NBI - Etablierungsphase-European Molecular Biology Laboratory-HD-HUB, G965081974
OpenAlex ID: W3143324940
