A compositional mediation model for a binary outcome: Application to microbiome studies

MOTIVATION The delicate balance of the microbiome is implicated in our health and is shaped by external factors, such as diet and xenobiotics. Therefore, understanding the role of the microbiome in linking external factors and our health conditions is crucial to translate the microbiome research into therapeutic and preventative applications. RESULTS We introduced a sparse compositional mediation model for binary outcomes to estimate and test the mediation effects of the microbiome utilizing the compositional algebra defined in the simplex space and a linear zero-sum constraint on probit regression coefficients. For this model with the standard causal assumptions, we showed that both the causal direct and indirect effects are identifiable. We further developed a method for sensitivity analysis for the assumption of the no unmeasured confounding effects between the mediator and the outcome. We conducted extensive simulation studies to assess the performance of the proposed method and applied it to real microbiome data to study mediation effects of the microbiome on linking fat intake to overweight/obesity. AVAILABILITY An R package can be downloaded from https://github.com/mbsohn/cmmb. SUPPLEMENTARY INFORMATION Supplementary files are available at Bioinformatics online.

[1]  J. Bressan,et al.  Potential mechanisms for the emerging link between obesity and increased intestinal permeability. , 2012, Nutrition research.

[2]  Hongzhe Li.  Microbiome, Metagenomics, and High-Dimensional Compositional Data Analysis , 2015 .

[3]  Yen-Tsung Huang,et al.  Hypothesis test of mediation effect in causal mediation model with high‐dimensional continuous mediators , 2016, Biometrics.

[4]  Eric Z. Chen,et al.  Inflammation, Antibiotics, and Diet as Environmental Stressors of the Gut Microbiome in Pediatric Crohn's Disease. , 2015, Cell host & microbe.

[5]  L. Keele,et al.  A General Approach to Causal Mediation Analysis , 2010, Psychological methods.

[6]  Hongzhe Li,et al.  Variable selection in regression with compositional covariates , 2014 .

[7]  D. Rubin,et al.  Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction , 2016 .

[8]  Paul Wilmes,et al.  Human Gut Microbiome: Function Matters. , 2017, Trends in microbiology.

[9]  Stijn Vansteelandt,et al.  Odds ratios for mediation analysis for a dichotomous outcome. , 2010, American journal of epidemiology.

[10]  P. Guttorp,et al.  Statistical Interpretation of Species Composition , 2001 .

[11]  S. West,et al.  A comparison of methods to test mediation and other intervening variable effects. , 2002, Psychological methods.

[12]  Hongzhe Li,et al.  Generalized linear models with linear constraints for microbiome compositional data , 2018, Biometrics.

[13]  P. Green.  Iteratively reweighted least squares for maximum likelihood estimation , 1984 .

[14]  Hongzhe Li,et al.  Compositional mediation analysis for microbiome studies , 2019, The Annals of Applied Statistics.

[15]  J. Aitchison,et al.  Log contrast models for experiments with mixtures , 1984 .

[16]  F. Bushman,et al.  Linking Long-Term Dietary Patterns with Gut Microbial Enterotypes , 2011, Science.

[17]  T J VanderWeele,et al.  Mediation Analysis with Multiple Mediators , 2014, Epidemiologic methods.

[18]  Jun Chen,et al.  Mediation effect selection in high‐dimensional and compositional microbiome data , 2020, Statistics in medicine.

[19]  A. Kurilshikov,et al.  Host Genetics and Gut Microbiome: Challenges and Perspectives. , 2017, Trends in immunology.

[20]  L. Keele,et al.  Identification, Inference and Sensitivity Analysis for Causal Mediation Effects , 2010, 1011.1079.

[21]  Ian D. Caterson,et al.  Increased Gut Permeability and Microbiota Change Associate with Mesenteric Fat Inflammation and Metabolic Dysfunction in Diet-Induced Obese Mice , 2012, PloS one.

[22]  Huilin Li,et al.  Estimating and testing the microbial causal mediation effect with high-dimensional and compositional microbiome data , 2019, bioRxiv.

[23]  K. Honda,et al.  The microbiota in adaptive immune homeostasis and disease , 2016, Nature.

[24]  Anru R. Zhang,et al.  Regression Analysis for Microbiome Compositional Data , 2016, 1603.00974.

[25]  D. A. Kenny,et al.  The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. , 1986, Journal of personality and social psychology.

[26]  N. Zmora,et al.  The microbiome and innate immunity , 2016, Nature.

[27]  Kosuke Imai,et al.  Identification and Sensitivity Analysis for Multiple Causal Mechanisms: Revisiting Evidence from Framing Experiments , 2013, Political Analysis.

[28]  D. Rubin.  Causal Inference Using Potential Outcomes , 2005 .