Mediation Analysis with Compositional Exposures
- 1Fred Hutchinson Cancer Center
Abstract
Understanding the causal pathways linking the gut microbiome to downstream biomarkers and clinical outcomes is central to microbiome research. When the microbiome acts as the exposure, the high dimensionality and compositional structure of the data introduce unique methodological challenges for mediation analysis. To address these challenges, we propose a latent variable mediation framework that captures variation in microbial composition through a microbial balance, defined as the log-ratio between two unknown subsets of taxa. This balance serves as a latent scalar exposure and simplifies the estimation of the overall indirect effect at the community level, while simultaneously identifying specific taxa that contribute to the overall direct and indirect effects. We describe the model's estimation and inference, and illustrate the method using data from the Multiethnic Cohort--Adiposity Phenotype Study, where we examine the mediation pathway from the gut microbiome through lipopolysaccharide-binding protein, a marker of metabolic endotoxemia, to percent liver fat.
Related
- Human macrophages encode stimulus-specific information of prior exposures through trained immunity
- Chlamydia trachomatis Incidence in Relation to Vaginal Microbiota Dynamics, Immunogenetics and Exposures in a Cohort of Young Student Women in France
- The compositional encoding of hand-eye coordinated movements for single neurons in the posterior parietal cortex
- Host Identity and Urban Space Structure Phyllosphere Bacterial Communities Through Compositional Turnover in a Tropical Andean City
- Neural basis of compositional control
BibTeX
@misc{ma2026mediation,
title = {{Mediation Analysis with Compositional Exposures}},
author = {J. Ma and P. Kundu and T. W. Randolph and S. L. Navarro and M. A. J. Hullar},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.24.754147},
url = {https://www.biorxiv.org/content/10.64898/2026.09.24.754147v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Mediation Analysis with Compositional Exposures
AU - J. Ma
AU - P. Kundu
AU - T. W. Randolph
AU - S. L. Navarro
AU - M. A. J. Hullar
PY - 2026
DA - 2026/09/24
PB - bioRxiv
DO - 10.64898/2026.09.24.754147
UR - https://www.biorxiv.org/content/10.64898/2026.09.24.754147v1
AB - Understanding the causal pathways linking the gut microbiome to downstream biomarkers and clinical outcomes is central to microbiome research. When the microbiome acts as the exposure, the high dimensionality and compositional structure of the data introduce unique methodological challenges for mediation analysis. To address these challenges, we propose a latent variable mediation framework that captures variation in microbial composition through a microbial balance, defined as the log-ratio between two unknown subsets of taxa. This balance serves as a latent scalar exposure and simplifies the estimation of the overall indirect effect at the community level, while simultaneously identifying specific taxa that contribute to the overall direct and indirect effects. We describe the model's estimation and inference, and illustrate the method using data from the Multiethnic Cohort--Adiposity Phenotype Study, where we examine the mediation pathway from the gut microbiome through lipopolysaccharide-binding protein, a marker of metabolic endotoxemia, to percent liver fat.
ER -