Korean Variant Archive (KOVA): a reference database of genetic variations in the Korean population

Abstract

Previous efforts in identifying and understanding human genome variations have been systematically preferred towards the populations of Caucasian descendants and have resulted in loss of accuracy and detection power when analyzing genome variations in non-Caucasian populations. To improve accuracy and precision in studying genome variations in the Korean population, we have compiled and curated a large-scale database of coding variants from 1,055 healthy Korean samples. These samples were sequenced to an average depth of 75x with 101 singleton variants per individual. Through analysis of our healthy Korean cohort, we found distinct ethnic group characteristics comparable to that of other ethnic groups in Africa and Europe, indicating that the Korean population are a discrete ethnic group that can benefit from an independent reference database. Through population genetics analyses, we found that our Korean Variant Archive (KOVA) increased variant filtering power when examining Korean exomes in concert with Exome Aggregation Consortium. Interestingly, we provide a cohort of potential germline variants that may be associated with susceptibility to tumorigenesis. These rare germline variants were validated using independent datasets from The Cancer Genome Atlas and 1000 Genome Project. To our knowledge, KOVA represents the first database curated using high quality whole exome sequencing-based variants specifically from Korean individuals that can serve as a valuable resource for future studies in the study of human genome variations in the Korean population.