Jing Ma

The paper introduces a latent variable mediation framework to analyze the gut microbiome's indirect effects on clinical outcomes through compositional data structures.

This paper constructs the first comorbidity networks in companion dogs using data from the Dog Aging Project, revealing both well-known and novel disease associations and their temporal relationships.

The paper introduces REHE and reREHE methods for fast and robust variance components estimation in linear mixed models, providing non-negative estimates comparable to REML.

The paper introduces a censored Gaussian graphical model to estimate microbial-metabolomic networks and demonstrates its effectiveness on simulated and bacterial vaginosis data.

This study compares nine pathway enrichment methods, finding that methods using both differential expression and pathway topology perform better on metabolomics data.

The paper introduces the GMD-biplot, a novel visualization method for microbiome data that improves clustering and variable-sample relationships compared to traditional PCoA plots.