Inverse probability weighting-based mediation analysis for microbiome data
Open Access
- 2 October 2026
- journal article
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 42 (10)
Abstract
Motivation: Mediation analysis is an important tool for elucidating causal mechanisms in biomedical research and has recently gained increasing attention in microbiome studies. Motivated by a microbiome study of acute myeloid leukemia (AML) patients, we investigate whether the effect of induction chemotherapy (IC) intensity on infection status is mediated through gut microbial taxa. This problem poses substantial methodological challenges due to the high dimensionality, zero inflation, and dependence of microbial mediators, as well as the presence of an exposure-induced mediator–outcome confounder, antibiotic use. Results: To address these challenges, we propose a novel nonparametric identification formula for the interventional indirect effect, a causal mediation estimand that remains identifiable in the presence of exposure-induced mediator–outcome confounding. Building on this identification result, we develop an inverse probability-weighting estimator and construct standard normal bootstrap confidence intervals for inference. Simulation studies demonstrate that the proposed method achieves accurate effect estimation and maintains appropriate type I error rates and power in finite samples. Application to the AML microbiome study provides evidence of a statistically significant interventional indirect effect of IC intensity on infection risk through alterations in the gut microbiome. Availability and implementation: The R code for the simulation studies is available at https://github.com/YX-ZHANG-1/CausalMediationMicrobiome and archived on Zenodo at https://doi.org/10.5281/zenodo.21960802Keywords
All Related Versions
Funding Information
- National Institutes of Health (P30 CA013696)
- National Institutes of Health (R01 AI143886)
- National Institutes of Health (R01 CA219896)
- National Institutes of Health (R01 CA256977)
- National Institutes of Health (P30 CA016672)
- Cancer Prevention and Research Institute of Texas (RP200633)
- Natural Sciences and Engineering Research Council of Canada (RGPIN-2019-07052)
- Natural Sciences and Engineering Research Council of Canada (RGPAS-2019-00093)
- Natural Sciences and Engineering Research Council of Canada (DGECR-2019-00453)
This publication has 40 references indexed in Scilit:
- Mediation effect selection in high‐dimensional and compositional microbiome dataStatistics in Medicine, 2020
- An Approach to Analyze Longitudinal Zero-Inflated Microbiome Count Data Using Two-Stage Mixed Effects ModelsStatistics in Biosciences, 2020
- The impact of early-life sub-therapeutic antibiotic treatment (STAT) on excessive weight is robust despite transfer of intestinal microbesThe ISME Journal, 2019
- Dynamics of the human gut microbiome in inflammatory bowel diseaseNature Microbiology, 2017
- The role of the gastrointestinal microbiome in infectious complications during induction chemotherapy for acute myeloid leukemiaCancer, 2016
- Richness of human gut microbiome correlates with metabolic markersNature, 2013
- Intestinal microbiota metabolism of l-carnitine, a nutrient in red meat, promotes atherosclerosisNature Medicine, 2013
- Recent changes in bacteremia in patients with cancer: a systematic review of epidemiology and antibiotic resistanceEuropean Journal of Clinical Microbiology & Infectious Diseases, 2013
- Infection-control interventions for cancer patients after chemotherapy: a systematic review and meta-analysisThe Lancet Infectious Diseases, 2009
- The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations.Journal of Personality and Social Psychology, 1986