Our lab's primary research interest lies in developing statistical/computational methods that accurately analyze bio big datasets by efficiently removing various confounding factors. My research interest can be categorized into several branches. The first branch focuses on eQTL (expression quantitative trait loci) studies. I have developed a new statistical method that accurately identifies regulatory hotspots while effectively capturing expression heterogeneity caused by various unknown confounding factors in eQTL studies. Further, I participated in many eQTL studies, in which I analyzed numerous eQTL datasets. The second branch of my research involves the development of genome-wide association study (GWAS) methods that accurately identify true signals by correcting population structure. Population structure is one of the widespread confounding factors in GWAS that complicates the analysis. I discovered that population structure cause serious problems in multiple phenotype analysis and developed a method that efficiently and accurately performs multiple phenotype analysis accounting for population structure. In addition, I developed a method that corrects for multiple hypothesis testing, accounting for population structure. Besides, I am interested in various other problems of GWAS, including privacy preserving protocols for identifying genetic relatives, fine mapping, meta analysis, heritability estimation, causal inference of gene regulation.