Vcfanno: fast, flexible annotation of genetic variants

Abstract

BackgroundThe integration of genome annotations and reference databases is critical to the identification of genetic variants that may be of interest in studies of disease or other traits. However, comprehensive variant annotation with diverse file formats is difficult with existing methods.\n\nResultsWe have developed vcfanno as a flexible toolset that simplifies the annotation of genetic variants in VCF format. Vcfanno can extract and summarize multiple attributes from one or more annotation files and append the resulting annotations to the INFO field of the original VCF file. Vcfanno also integrates the lua scripting language so that users can easily develop custom annotations and metrics. By leveraging a new parallel \"chromosome sweeping\" algorithm, it enables rapid annotation of both whole-exome and whole-genome datasets. We demonstrate this performance by annotating over 85.3 million variants in less than 17 minutes (>85,000 variants per second) with 50 attributes from 17 commonly used genome annotation resources.\n\nConclusionsVcfanno is a flexible software package that provides researchers with the ability to annotate genetic variation with a wide range of datasets and reference databases in diverse genomic formats.\n\nAvailabilityThe vcfanno source code is available at https://github.com/brentp/vcfanno under the MIT license, and platform-specific binaries are available at https://github.com/brentp/vcfanno/releases. Detailed documentation is available at http://brentp.github.io/vcfanno/, and the code underlying the analyses presented can be found at https://github.com/brentp/vcfanno/tree/master/scripts/paper.