SNV Frequencies Korea KOVA 5.3k mixed Track Settings
Version: V7 Assembly: Human Dec. 2013 (GRCh38/hg38)
Description
The Korean Variant Archive (KOVA)
contains 1,896 whole genome sequencing and 3,409 whole exome sequencing data from healthy
individuals of Korean ethnicity. Most of the samples originated from normal tissue of cancer
patients (40.16%), healthy parents of rare disease patients (28.4%), or healthy volunteers
(31.44%). Korean ancestry is not broken down further in the INFO field. Coverage 100x for WES, 30x for WGS.
SVs called with Manta are also available.
Data Access
Due to license restrictions, the data for this track cannot be downloaded from the UCSC
Genome Browser. The Table Browser, Data Integrator, and download server are not available
for this track.
TSV data can be requested on the KOVA Downloads website. Our GitHub repo contains a script that
converts this format to VCF.
Methods
Raw reads were aligned to the GRCh38+decoy reference with BWA-MEM v0.7.17 with default
parameters. Duplicates were marked and coordinates sorted with MarkDuplicatesSpark, then base
quality scores were recalibrated with BQSRPipelineSpark in GATK v4.1.3.0. Mapping quality control
metrics were generated with Qualimap v2.2.1. Single-nucleotide variants and small
insertions/deletions were called per sample with GATK HaplotypeCaller in GVCF mode (-ERC GVCF), and
joint genotyping was performed by creating a GenomicsDB with GenomicsDBImport and followed GATK
Best Practices. Variant quality score recalibration (VQSR) retained 99.7% of true SNVs
and 99.0% of true indels based on training sets (workflow detailed in Supplementary Fig. 1).
Downstream analyses followed a modified version of the gnomAD quality-control framework and were
primarily conducted with Hail. After WES and WGS data were merged in Hail, multiallelic variants and
variants with genotype quality <20, read depth <10, allelic balance <0.2, or overlap with
low-complexity regions were excluded.
At UCSC, V7 of the TSV.gz was obtained from the KOVA staff by email and converted to VCF. The file is not
available for download from our site, but can be requested from the KOVA website.
The makeDoc file for the varFreqs track documents how all source files were converted.
Python scripts used for some of the tracks are available on GitHub.
Credits
Thanks to Insu Jang and the KOVA director for providing variant frequencies in TSV format.