Wonil Chung

I am a research associate at Harvard T.H. Chan School of Public Health. I received Ph.D. degree in Biostatistics from the University of North Carolina at Chapel Hill (UNC) and MS, BS in Statistics from Seoul National University (SNU) in Korea.  I believe that statistical interpretation would be truly crucial to understand various life phenomena, especially genetic phenomena, therefore the development of statistical methods for genomic data would greatly help us to discover the mystery of life.

As my early career, I worked for an IT company as a computer programmer and thus have strong programming skill in C/C++, Java and R, which is beneficial for implementing statistical methods and analyzing genetic data. During my doctoral and postdoctoral years, I have developed novel statistical methodologies for genome-wide association studies (GWAS), quantitative trait loci (QTL), expression QTL (eQTL) mapping and genomic risk prediction. Also, I have participated in a variety of large-scale omics projects such as identification of shared genetic architecture and analyses of methylation and metabolomics data.

Specifically, for my doctoral dissertation, I developed Bayesian parametric and nonparametric methods for multiple QTL mapping. I extended existing Bayesian parametric and nonparametric methods from univariate traits to longitudinal traits for mapping multiple QTL. In order to estimate heritability and conduct eQTL analysis using twin data, I developed an efficient REML mixed model using variance-component maximization. To predict gene expression into GWAS cohorts, I participated in the development of the methodologies for transcriptome-wide association studies (TWAS). To improve prediction accuracy of phenotypes of interest using related other phenotypes, I developed cross-trait penalized regression (CTPR), a powerful and practical approach for multi-trait polygenic risk prediction in large cohorts . The proposed CTPR outperformed other existing multi-trait prediction methods. Lastly, I analyzed large-sample GWAS data to identify shared genetic architecture and related tissues for a wide range of complex traits and diseases.

Wonil Chung, Ph.D.
Research Associate
Harvard T.H. Chan School of Public Health (HSPH)
655 Huntington Ave. Building II Room 202, Boston, MA 02115
wchung(at)hsph.harvard.edu
https://www.hsph.harvard.edu/wonil-chung/