Methods for Mediation Analysis with High-Dimensional DNA Methylation Data: Possible Choices and Comparison

Abstract

Epigenetic researchers often evaluate DNA methylation as a mediator between social/environmental exposures and disease, but modern statistical methods for jointly evaluating many mediators have not been widely adopted. We compare seven methods for high-dimensional mediation analysis with continuous outcomes through both diverse simulations and analysis of DNAm data from a large national cohort in the United States, while providing an R package for their implementation. Among the considered choices, the best-performing methods for detecting active mediators in simulations are the Bayesian sparse linear mixed model by Song et al. (2020) and high-dimensional mediation analysis by Gao et al. (2019); while the superior methods for estimating the global mediation effect are high-dimensional linear mediation analysis by Zhou et al. (2021) and principal component mediation analysis by Huang and Pan (2016). We provide guidelines for epigenetic researchers on choosing the best method in practice and offer suggestions for future methodological development.

Network analysis identifies strain-dependent response to tau and tau seeding-associated genes

“…First, high-dimensional mediation analysis ( Zhang et al, 2016 ) was performed using R Package for High-Dimensional Mediation Analysis ( http://github.com/dclarkboucher/hdmed ; Clark-Boucher et al, 2023 Preprint ). Alpha, beta, and P values were calculated using the default parameters of the mediate_hima() function where A = genetic background, M = matrix containing the eigengene value of each WGCNA module, and Y = tau seeding measured by FRET.…”

The murine lung microbiome is dysbalanced by the human pathogenic fungusAspergillus fumigatusresulting in enrichment of anaerobic bacteria

“…The correlation value was calculated using the RV2 method of the R package MatrixCorrelation 74 , and the p-value was obtained using a permutation test. The mediation analysis was performed using the hdmed (v 1.0.0, HIMA method) package in R 75,76 and its mutant derivatives ∆sidA 81 and ∆freB/∆sidA 82 . A. fumigatus was cultured in AMM and on AMM agar plates.…”

multimedia: Multimodal Mediation Analysis of Microbiome Data

“…1 A represents this graphically, with separate paths for treatment T → mediator M → outcome Y and treatment T → outcome Y . In the causal inference literature, this exercise is called mediation analysis, and various techniques have emerged to support it [29, 8]. Several adaptations have been proposed for the microbiome setting, where mediators, outcomes, and controls may be high-dimensional [38, 48, 5, 23].…”