Xiaoyu Lu

This paper introduces Somatic-IV, a method to identify candidate driver genes in cancer by integrating genetic, transcriptomic, and clinical data, and finds ZBED2 and CSNK2A2 as potential drivers in pancreatic and colorectal cancers.

The paper uses computational modeling and transcriptomic data to show that cancer cell phenotypes can be explained by Fenton reaction-driven proton production and reprogrammed metabolisms.

This paper introduces scFEA, a graph neural network model to estimate cell-wise metabolic flux from scRNA-seq data, validated with matched metabolomics data and applied to multiple datasets.

SSMD improves cell type identification and deconvolution in mouse transcriptomics by addressing variations in genetics, experimental platforms, and small sample sizes.

ICTD improves tissue microenvironment analysis by enhancing cell type identification, marker gene discovery, and functional inference through a semi-supervised deconvolution approach.

The paper introduces LTMG, a model for capturing transcriptional regulatory heterogeneities in scRNA-seq data, showing improved fitting and validation on multiple datasets.