Ruixin Zhu

This paper benchmarks confounder-aware differential abundance methods for microbiome studies, finding that covariate adjustment improves false discovery control but reduces sensitivity, with different strategies performing better under various conditions.

IBDkb integrates multi-source IBD data with AI tools to enhance research and drug discovery by enabling efficient exploration of drug-target-trial associations.

DeepSeMS uses a large language model to predict chemical structures of secondary metabolites from microbial gene clusters, outperforming existing methods and discovering novel compounds in the ocean microbiome.

The paper systematically identifies secondary bile acid production genes across microbial genomes and highlights their associations with diseases.

The paper compares microbiota and metabolome profiles of quiescent Crohn's disease patients with their healthy first-degree relatives to identify biomarkers associated with disease relapse.

The paper identifies eight keystone species in NAFLD gut microbiota using causal inference and dynamic intervention modeling, which can restore microbial composition toward a normal state.

This study develops a drug repositioning platform by integrating anti-fibrosis knowledge and machine learning models to identify potential drugs for fibrosis treatment.