Xiaochen Yang

Domain-grounded prompting lets GPT-5 produce genetic evidence for 13,022 drug target-indication pairs that predicts clinical success as well as expert curation, with wider coverage.

The paper maps phenotypic and genetic links between plasma proteins and organ imaging traits using UK Biobank data, identifying associations and causal links across multiple organs.

This study uses Mendelian randomization to identify genetic causal links between 58 diseases, 56 imaging traits, and multiple organs across 402 imaging traits and 372 clinical outcomes.

The paper investigates organ-specific and modality-specific aging clocks using proteomic and imaging data, revealing distinct biological processes and their roles in age-related diseases.

PennPRS provides a cloud platform for training polygenic risk scores without individual-level data, validated with simulations and real data across 6,000 phenotypes.

The paper introduces BLISS, a framework for training proteome-wide association models using pQTL summary statistics, enabling multi-ancestry analysis of protein-phenotype associations.

The paper introduces BIGA, a cloud platform for analyzing cross-trait genetic architecture using GWAS summary statistics, enabling unified data analysis and resource sharing.

The paper uses genetic data to predict imaging-derived phenotypes from the UK Biobank, creating polygenic scores that identify disease risk in external cohorts.

The paper identifies 60 genetic loci associated with brain ventricular and subcortical shape features using shape measures from over 45,000 UK Biobank and ABCD study subjects.

This study identifies genetic overlaps between sleep traits and multi-organ imaging biomarkers, showing strong genetic links between sleep and psychiatric disorders.

This study identifies genetic loci associated with functional brain networks and areas using large-scale fMRI data, linking them to cognitive traits and brain disorders.