Associate Professor, Public Health Sciences
Statistical genetics and genomics.
My research focuses on the design and statistical analysis of human gene mapping data. Recently, my research has focused on the development of methods for analyzing genome-wide SNP data in datasets that include thousands of individuals.
Genome Wide Association Analysis
I developed a general procedure to infer missing genotypes in family-based genomewide association studies (Chen and Abecasis 2007). This work allows efficient use of family collections in genome wide association (GWA) studies, especially when the GWA follows an initial linkage scan. With this approach, only a few selected individuals in each family need to be genotyped during the GWA, and genotypes of the remaining individuals can be mostly inferred based on the information from flanking markers and relatives. In this work, I also proposed efficient tests for GWA analysis of quantitative traits that allow for uncertainty in the estimation of missing genotypes. These methodologies have been successfully applied to the genomewide association analysis of the gene expression data as well as a large GWA study of ~100 aging related quantitative traits (mainly cardiovascular and personality traits) in >6,000 individuals from Sardinia.
Robust Linkage and Association Tests
I developed a general methodology framework for quantitative trait linkage analysis making use of the Generalized Estimating Equations (GEE) (Chen et al. 2004). Using this framework, I proposed novel robust linkage tests and investigated commonly used linkage methods (Chen et al. 2005).
Power Analysis of Linkage and Association
I developed an efficient method to perform analytical power calculation for the variance component linkage analysis in pedigrees of any size (Chen and Abecasis 2006). I also developed a general approach to calculate the power of the Transmission/Disequilibrium Test (TDT), a widely-used test and design for association studies (Chen and Deng 2001).