Fuzhong Xue

This paper develops a federated framework for GWAS using mixed-effects models to enable privacy-preserving analysis across distributed datasets while maintaining statistical accuracy.

JointMR integrates GWAS summary databases to estimate causal effects, providing unbiased and reliable results with higher power and better Type I error control than traditional methods.

TEMR improves causal effect estimation in non-European populations using trans-ethnic GWAS data, identifying 13 new causal relationships.

MRSL uses GWAS summary data to discover causal relationships between biomarkers and diseases, achieving higher accuracy and speed than existing methods.

This paper evaluates GWAS adjustment strategies and recommends adjusting for SNPs in linkage disequilibrium rather than demographic or clinical covariates.

The paper introduces SeRenDIP-CE, a sequence-based method for predicting conformational epitopes, achieving better performance than existing predictors on an antibody-antigen structure.

The paper uses machine learning to predict brain age from diffusion tensor imaging in schizophrenia patients and finds accelerated white matter aging is associated with cognitive deficits.

The paper introduces a semi-parametric latent differential network model for analyzing genetic networks across conditions, showing it works well for various data types and provides deeper insights into disease mechanisms.

JDINAC identifies differential interaction networks and classifies diseases using high-dimensional sparse omics data without parametric assumptions, achieving accurate results in breast cancer analysis.