Xuekui Zhang

Systematic benchmarking reveals pseudo-bulk biases and provides practical fixes for accurate RNA deconvolution.

The paper constructs a standardized BAL scRNA-seq atlas using BAL-EA, a novel annotation framework, revealing disease-associated shifts in COPD and validating findings in COVID-19 datasets.

MOH improves single-cell clustering by integrating three omics types into a multilayer heterogeneous graph, achieving better performance than existing methods.

scAGCI improves cell clustering in integrated scRNA-seq and scATAC-seq data by using anchor graphs to capture shared and specific information, achieving better performance and biological relevance.

scAnnotate is a cell type annotation tool for scRNA-seq data that uses dropout information through a mixture model and ensemble learning, showing competitive performance across multiple datasets.

cSurvival is a web resource that enables biomarker interaction analysis and integrates clinical and experimental cancer data to improve prognosis and drug resistance insights.

The paper extends a SARS-CoV-2-human protein-protein interaction network using a probabilistic database and identifies dense subgraphs to generate novel hypotheses about virus-targeting genes, proteins, and diseases.