The variant call format provides efficient and robust storage of GWAS summary statistics

Abstract

GWAS summary statistics are fundamental for a variety of research applications yet no common storage format has been widely adopted. Existing tabular formats ambiguously or incompletely store information about genetic variants and associations, lack essential metadata and are typically not indexed yielding poor query performance and increasing the possibility of errors in data interpretation and post-GWAS analyses. To address these issues, we adapted the variant call format to store GWAS summary statistics (GWAS-VCF) and developed open-source tools to use this format in downstream analyses. We provide open access to over 10,000 complete GWAS summary datasets converted to this format ( https://gwas.mrcieu.ac.uk ).

Keywords: GWAS; Storage format; Summary statistics; VCF.

Fig. 1

Performance comparison for querying summary statistics in plain text and GWAS-VCF. Mean query time (seconds, lower is quicker; repetitions n = 100) to extract either a single variant using the chromosome position or dbSNP [31] identifier or multiple variants using a 1-Mb interval or association P value. AWK, grep, bcftools [23] and rsidx [32] were evaluated using uncompressed/GZIP compressed TSV and BGZIP [23] compressed VCF. The summary statistics files contained one (single) or five (multiple) GWAS studies which were prepared by subsampling variants (n = 0.5 M, 2.5 M, 10 M) obtain from Neale et al. [35]. Error bars represent the 95% confidence interval