Hon‐Cheong So

This paper identifies shared and differential genetic factors between COVID-19 and other respiratory disorders using multi-omics data and provides insights for targeted therapies.

This study conducts a GWAS and cross-ethnic meta-analysis of schizophrenia phenotypes in Hong Kong Chinese and European populations, identifying genetic variants and building predictive models for clinical outcomes.

This study identifies genetic variants associated with antipsychotic-induced lipid and BMI changes in schizophrenia patients using GWAS, offering potential for personalized treatment strategies.

This study uses CombGWAS to analyze genetic correlations between psychiatric disorder combinations and cardiovascular disease, identifying shared and unique genetic factors.

CausalCellInfer integrates single-cell RNA-seq with biobank-scale GWAS to identify cell-type-specific disease mechanisms, revealing known and novel biological signals in traits like Type 2 Diabetes and depression.

INSIDER is a statistical framework for analyzing bulk RNA expression data that models interactions between biological variables and decomposes variation into a shared latent space to improve interpretation and downstream analysis.

This study shows pathway-specific polygenic scores improve cross-ancestry prediction of psychosis risk and clinical outcomes compared to genome-wide polygenic scores.

SR2 learns sparse representations for scRNA-seq data integration, capturing gene expression patterns under heterogeneous conditions and identifying cell identities effectively.

This paper introduces a Bayesian network framework to estimate direct and indirect causal effects of genes on complex traits using GWAS data and gene expression, showing improved enrichment of known genes and pathways.

This study conducts a genome-wide association study on Chinese and English language abilities in bilingual Hong Kong children, identifying 9 significant genetic loci and showing polygenic overlap with educational attainment and ADHD.

This paper uses machine learning to predict disease-specific drug targets by integrating drug-induced expression profiles and gene perturbation data, finding significant enrichment of known targets.

This study identifies zonation-dependent transcriptomic changes in aged brain endothelial cells, highlighting functional impacts on the BBB and AD-associated genes, and shows reversal with GLP-1R agonist treatment.

The paper identifies a novel copy number variant associated with suicide attempts in major depressive disorder through genome-wide analysis and validation.

This study identifies causal relationships between plasma proteins and psychiatric disorders using proteome-wide Mendelian randomization, revealing potential biomarkers and drug targets.

This study uses Mendelian randomization to find that triglycerides causally link to depression symptoms and self-harm, while HDL cholesterol shows weaker associations.

This paper introduces a framework integrating GWAS-derived transcriptomes and clinical data to subtype complex diseases, showing improved clustering performance and biological relevance in schizophrenia and cardiovascular risk stratification.

This study explores the role of de novo mutations in neuropsychiatric disorders and identifies drug classes enriched for these mutations, suggesting potential for drug discovery and repositioning.

This study uses Mendelian randomization to investigate statin repositioning opportunities and side-effects across 55 traits, finding associations between LDL-C reduction and various outcomes.

This study uses GWAS gene-set analyses to reposition drugs for depression and anxiety, finding enrichment of psychiatric medications and new candidates like fendiline.

This study identifies shared genetic factors between depression subtypes and cardiometabolic risks, suggesting subtype-specific associations and potential drug repositioning opportunities.

This study repositions drugs for seven psychiatric disorders by comparing GWAS-imputed transcriptomes with drug expression profiles from the Connectivity Map, identifying novel candidates supported by preclinical or clinical evidence.