Yulin Dai

PerturbLDM predicts single-cell transcriptional responses to perturbations using a latent diffusion model, outperforming existing methods in drug-dose-cell line combinations and capturing key biological programs.

GRPa-PRS identifies genetically-regulated pathways in polygenic diseases by stratifying individuals based on PRS and analyzing differential gene expression across risk strata.

scEMB is a transformer-based model that learns gene embeddings from single-cell transcriptomics data, outperforming existing models in tasks like clustering and cell type annotation.

deCS improves cell type annotation in scRNA-seq data by using a comprehensive human cell type reference atlas, enhancing accuracy and reducing computation time.

The paper identifies CXCR6 as a host genetic factor associated with severe COVID-19 through transcriptomic analysis and shows its protective effect in lung tissue.

TSEA identifies tissue-specific genes and enriches query data by analyzing multi-trait genome-wide association and cancer expression data.

This paper introduces Gene2Vec, a gene embedding method using co-expression data from GEO databases, showing improved pathway recovery and gene interaction prediction.