Machine learning approaches in microbiome research: challenges and best practices.

Microbiome data predictive analysis within a machine learning (ML) workflow presents numerous domain-specific challenges involving preprocessing, feature selection, predictive modeling, performance estimation, model interpretation, and the extraction of biological information from the results. To assist decision-making, we offer a set of recommendations on algorithm selection, pipeline creation and evaluation, stemming from the COST Action ML4Microbiome. We compared the suggested approaches on a multi-cohort shotgun metagenomics dataset of colorectal cancer patients, focusing on their performance in disease diagnosis and biomarker discovery. It is demonstrated that the use of compositional transformations and filtering methods as part of data preprocessing does not always improve the predictive performance of a model. In contrast, the multivariate feature selection, such as the Statistically Equivalent Signatures algorithm, was effective in reducing the classification error. When validated on a separate test dataset, this algorithm in combination with random forest modeling, provided the most accurate performance estimates. Lastly, we showed how linear modeling by logistic regression coupled with visualization techniques such as Individual Conditional Expectation (ICE) plots can yield interpretable results and offer biological insights. These findings are significant for clinicians and non-experts alike in translational applications.

A microbiota-modulated checkpoint directs immunosuppressive intestinal T cells into cancers.

Antibiotics (ABX) compromise the efficacy of programmed cell death protein 1 (PD-1) blockade in cancer patients, but the mechanisms underlying their immunosuppressive effects remain unknown. By inducing the down-regulation of mucosal addressin cell adhesion molecule 1 (MAdCAM-1) in the ileum, post-ABX gut recolonization by Enterocloster species drove the emigration of enterotropic α4ß7+CD4+ regulatory T 17 cells into the tumor. These deleterious ABX effects were mimicked by oral gavage of Enterocloster species, by genetic deficiency, or by antibody-mediated neutralization of MAdCAM-1 and its receptor, α4ß7 integrin. By contrast, fecal microbiota transplantation or interleukin-17A neutralization prevented ABX-induced immunosuppression. In independent lung, kidney, and bladder cancer patient cohorts, low serum levels of soluble MAdCAM-1 had a negative prognostic impact. Thus, the MAdCAM-1-α4ß7 axis constitutes an actionable gut immune checkpoint in cancer immunosurveillance.