Suying Bao

The paper improves rare variant association studies by using proteomics-informed large language models, identifying more gene-trait associations than existing methods.

DeltaSplice predicts splicing-altering mutations and alternative splicing from genomic sequences, outperforming existing methods and identifying disease-related variants in neurodevelopmental disorders.

This paper presents the largest catalog of human protein-coding variation from 985,830 individuals, identifying rare variants and their functional implications for precision medicine.

The paper identifies rare coding variants in CHRNB2 associated with reduced smoking likelihood and links them to downstream smoking-related phenotypes.

The paper identifies a rare variant that lowers ACE2 expression and reduces COVID-19 risk, and develops a genetic risk score predicting severe disease.

The paper develops a computational method, mCross, to model RBP binding specificity using precise crosslink site registration and validates it with ENCODE eCLIP data.