Yi Liu

The paper introduces π-MSNet, a large-scale proteomics dataset for AI training and benchmarking, showing performance improvements when models are retrained on it.

This study uses large biobanks to identify gene-trait pairs with causal evidence, rediscover approved drug targets, and develop a predictive model for drug development.

DeepRTAlign improves retention time alignment for large cohort LC-MS data, enhancing identification sensitivity without compromising quantitative accuracy and enabling predictive modeling for hepatocellular carcinoma recurrence.

This paper uses Mendelian randomization and electronic health records to show SGLT2 inhibitors reduce prostate cancer risk, with additional evidence from differential expression analysis.

The paper maps human traits from UK Biobank to the Experimental Factor Ontology using language models, finding BioSentVec and BlueBERT-EFO perform best.

This paper uses proteome-wide Mendelian randomization across multiple ancestries to identify causal protein-disease associations and drug targets in over 1.26 million individuals.

This study uses multi-ancestry Mendelian randomization to identify six protein targets for COVID-19 severity, showing ancestry-specific effects.

This paper develops a deep learning model to quantify cognitive reserve using structural MRI data from healthy cohorts and validates its generalizability across different populations and scanners using transfer learning.

EpiGraphDB is a graph database and analytical platform for health data science, demonstrating its utility through case studies on pleiotropy, drug targets, and causal inference.

This paper uses Mendelian randomization and colocalization to identify causal relationships between plasma proteins and disease phenotypes, finding 111 putatively causal effects.

OpenGWAS serves 14582 harmonised, complete GWAS summary datasets (125.8 billion associations) via API, R/Python clients and GWAS-VCF downloads, with glue code for Mendelian randomization, colocalisation and fine-mapping.

This study shows that PPARD promotes colorectal cancer progression and invasion by enhancing β-catenin signaling and multiple pro-invasive pathways.

This study shows that PPARD overexpression in gastric progenitor cells drives gastric tumorigenesis in mice by promoting inflammation, self-renewal, and tumorigenic properties.