Default header image

Sample Page

Introduction

My research lab is working on a broad range of projects but share a common theme on the development of statistical and computational tools for analyzing diverse type of data and addressing issues arising from genetic and biological studies. I am also interested in modelling the spread of infectious disease, with emphasis on methodology development for improving the model fitting, efficiency, and addressing issue of missing information in data.

In statistical genetics, my research aims at finding genes responsible for complex traits using the genome-wide coverage of single nucleotide polymorphisms (SNPs). From thousands to millions of SNPs, we screen out the noises and obtain a small subset that are significantly associated with the traits. We develop different methods for such analysis when the data are collected using different study designs (e.g., longitudinal, cross-sectional, single trait, multiple traits, family-based, and population-based) and from different target populations (e.g., human and animals). Specifically, I am interested in the following projects:

  • Genetic association with longitudinal phenotypes when family data is analyzed.
  • Gene-environmental interaction and time-varying gene
  • Missing data in longitudinal studies
  • Haplotype based association analysis
  • Genotype imputation from a small panel to a large panel
  • Low-density panel design of proven pig selection for breeding in Canadian swine industry
  • Ultra low-density panel design for within litter piglet selection

In Bioinformatics, I am interested in using microbiome sequencing data, to study human microbiome composition and their association with human physiology. I am currently working on these projects:

  • Clustering microbiome data via Finite-mixture of Dirichlet-multinomial regression models
  • Finite-mixture of Dirichlet-multinomial(DM) regression models for longitudinal microbiome data
  • Variable selection problems in clustering microbiome data using finite mixture of DM regression models
  • Finite mixtures of DM regression models for unsupervised and semi-supervised clustering

All my research projects require strong computational skill and good statistics background. My lab currently has 4 Masters students, 4 Ph.D. students, and one postdoctoral fellow. I also supervise Masters students in Guelph’s Bioinformatics program. I regularly recruit Masters and Ph.D. students every year.