Bernard Ng

Background

About Bernard Ng

My research interest lies in developing statistical and machine learning methods for biomedical applications. My graduate work focused on graphical models, community detection, Bayesian statistics, and sparse model learning for analysis of functional magnetic resonance imaging (fMRI) data. During my postdoctoral training, I proposed a number of sparse Gaussian graphical models (SGGM) for brain graph estimation. I also pursued research in statistics on Riemannian manifolds and high dimensional statistical testing. Further, I expanded my research into genetics and genomics, where I conducted large scale molecular quantitative trait locus analyses. Continuing down this line of research as a research associate, I proposed a multi-omic integration method for identifying disease genes based on the regulatory effects of GWAS variants. Currently, I am exploring deep learning for predicting gene expression based on DNA sequence as well as developing models for associating genes to brain connectivity. Ultimately, my goal is to integrate genetics, multi-omic, imaging, and clinical data to analyze relationships across the genome, epigenome, transcriptome, brain connectome, and phenome.