Han Chen

The paper introduces sum-h2, a fast framework for evaluating genetic discovery from deep learning-derived phenotypes using heritability measures, and establishes a Genetic Discovery Arena for benchmarking.

TorchGWAS2 improves genome-wide association testing in related samples by using a deterministic variance-correction algorithm, achieving faster and more powerful results across large-scale omics data.

This study identifies host DNA repair factors that enable antiviral nucleoside analog resistance by facilitating viral DNA synthesis in the presence of ganciclovir.

The paper identifies genetic drivers of protein change over time using longitudinal proteomics data from MESA, UK Biobank, and CARDIA, finding that baseline adjustment improves replication but may introduce biases.

The paper constructs a multi-omics single-cell atlas of human RPE and choroid, identifying cell types, aging patterns, and genetic risk factors for AMD.

The paper creates a multi-omics atlas of human trabecular meshwork and ciliary body cells, identifying cell types and regulatory programs linked to glaucoma.

The paper presents an xQTL atlas of regulatory variation in the aging human brain, integrating multiple modalities and cell types across four studies, identifying more variant-molecular-trait associations by including TAD boundaries.

HiFiMAP provides high-resolution IBD mapping at SNP level, outperforming window-based methods and detecting novel associations in UK Biobank and brain imaging data.

The paper introduces the expected polygenic risk score (ePRS) framework to address population stratification in polygenic risk score analysis by modeling ancestral makeup, showing it provides equitable risk quantification across diverse populations.

The paper introduces ARG-RHE, a method to estimate heritability and perform association testing using ancestral recombination graphs, showing improved results over imputation alone.

The paper introduces a robust mixed model association test (RoM) for gene-environment interactions in related samples, showing improved type I error control compared to existing methods.

The paper analyzes circulating metabolites using genome-wide association data from the TOPMed program, identifying 1,778 loci and highlighting sex-specific genetic contributions.

The paper introduces MultiSTAAR, a statistical framework for multi-trait rare variant analysis in WGS studies, identifying new lipid trait associations in diverse populations.

The paper identifies gene-diet interactions for fish oil supplementation and PUFA levels using GWIS in UK Biobank data, finding significant loci around FADS and GRP12.

The paper introduces JAGWAS, a multi-phenotype GWAS tool that identifies more genetic loci for brain imaging phenotypes than single-phenotype GWAS.

This study identifies gene-environment interactions between excessive daytime sleepiness and genetic variants linked to obstructive sleep apnea severity, revealing novel therapeutic targets.

FedGMMAT enables privacy-preserving genetic association testing across institutions using federated learning and homomorphic encryption.

The paper identifies novel genetic loci associated with systolic blood pressure that interact with the DASH diet score through genome-wide interaction analysis.

The paper introduces two tools, REGEM and METAGEM, to analyze gene-environment interaction summary statistics from GWIS studies, enabling efficient re-analysis and meta-analysis.

StocSum generates stochastic summary statistics for whole genome sequencing studies without external reference panels, enabling various genetic analyses and improving utility for underrepresented populations.

The paper identifies a gene-diet interaction modifying the association between macronutrient intake and glycemic traits, highlighting challenges in detecting such interactions due to measurement error and sample size.

This paper introduces an unsupervised deep learning method to derive brain imaging phenotypes from MRI data, identifying new genetic loci associated with these endophenotypes.

Monopogen calls single nucleotide variants from single-cell sequencing data, achieving high accuracy in germline variant detection and enabling ancestry inference and variant association analysis.

The paper introduces a framework for detecting noncoding rare variant associations using whole-genome sequencing data, applying it to lipid traits in the TOPMed program.

This paper uses gradient boosted trees to improve polygenic risk prediction by capturing non-linear genetic effects and SNP interactions, showing significant increases in variance explained for multiple complex traits.

The paper identifies variance-quantitative trait loci (vQTLs) and gene-environment interactions (GEIs) for cardiometabolic serum biomarkers using UK Biobank data.

The paper introduces iGWAS, a method using self-supervised deep learning to extract phenotypes from medical images for GWAS, finding new genetic loci related to retinal vessel development.

The paper introduces a method to generate summary statistics for gene-environment interaction tests and a meta-analysis approach to combine results across studies, demonstrating its effectiveness in large-scale sequencing data.

This paper develops a privacy-aware kinship inference method for admixed populations using reference panels and homomorphic encryption to protect genetic data.

This paper introduces FHAT and FHAT-O methods to incorporate family history in genetic association tests, showing improved power and type I error control in exome-wide analyses of dementia and hypertension.