Qi Zhang

MedDiC estimates high dimensional indirect effects using a difference-in-coefficients approach, showing improved power and speed over existing methods in simulations and biological applications.

The paper shows that prenatal stress in Cntnap2 heterozygous mice disrupts 5hmC levels and alters gene expression, leading to neurodevelopmental behaviors similar to knockout mice.

The paper develops MedFix and MedMix for high-dimensional mediation analysis, showing improved accuracy and robustness in causal gene identification using a mouse F2 diabetes dataset.

The paper applies genome-wide mediation analysis to maize data, identifying 736 mediating genes that explain 12.7% of phenotypic variance, with pleiotropic effects observed in 11% of genes.

This paper introduces EBHiC, an empirical Bayes method for peak detection in Hi-C data, showing improved accuracy and consistency compared to Fit-Hi-C.

This paper introduces TGCN, a method for analyzing temporal transcriptomic data using a low-rank plus sparse framework, improving covariance estimation and identifying robust gene modules.

This paper uses Bayesian confirmatory factor analysis and Bayesian networks to analyze rice phenotypes and identify genetic correlations among traits.