Zuoheng Wang

The paper evaluates estradiol's effects on ASD/NDD genes using human neurons and zebrafish, finding both convergent transcriptomic effects and gene-specific cellular and behavioral rescue.

GITIII infers cell-cell interactions from spatial transcriptomics data by analyzing cell state and niche correlations using a self-supervised graph transformer model.

This study identifies age-dependent DNA methylation associations with blood pressure using longitudinal data, revealing dynamic epigenetic changes linked to vascular aging.

This study identifies convergent biological pathways in mature neurons affected by neurodevelopmental disorder risk genes and shows drug candidates can ameliorate behavioral phenotypes.

This study introduces RVMMAT, a method for detecting time-varying genetic effects on longitudinal binary traits, and applies it to identify hypertension-associated loci.

This paper introduces TCGN, a deep learning model that integrates CNN, GNN, and transformer for gene expression prediction from histopathological images, achieving higher accuracy and interpretability.

This paper develops a longitudinal GWAS model with time-varying coefficients to detect genetic effects in Alzheimer's disease, identifying SNPs with age-dependent effects on clinical and biological phenotypes.

SDePER improves cell type deconvolution in spatial transcriptomic data by addressing platform effects, sparsity, and spatial correlation, achieving more accurate results than existing methods.

iDESC identifies differential expression in scRNA-seq data with multiple subjects by separating subject effects from disease effects, showing better performance than existing methods.

This study introduces two statistical tests to detect gene-environment interactions in hypertension, identifying two genes interacting with alcohol use.

G2S3 improves single-cell RNA sequencing data by imputing dropouts using a gene graph, enhancing expression recovery and analysis.

This paper introduces a framework to infer dynamic, personalized gene regulatory networks from transcriptional data and applies it to study surgical outcomes in a clinical genomic study.

This study introduces L-BRAT and RGMMAT for analyzing longitudinal binary traits, identifies loci and pathways related to cocaine use, and replicates findings in an independent GWAS.

Funmap2 is an R package for QTL mapping with longitudinal phenotypes, using a functional mapping framework that integrates biological knowledge and automatically selects models.