The presence of somatic copy-number alterations in tumor genomes can be used to predict both patient sensitivity to treatments as well as outcomes. The inclusion of allelic data improves statistical power to detect copy-number events and allows for discovery of copy-neutral events. We present GATK ACNV, an allelic copy-number variation method built on the Genome Analysis Toolkit. ACNV is a tool for detecting somatic copy-number activity from whole exome and whole genome sequencing data by segmenting the genome into regions of constant copy number and estimating copy ratio and minor-allele fraction in those regions.
ACNV uses a novel probabilistic model to account for reference bias (optionally using a panel of normals), which improves the estimation of minor-allele fraction. We combine this with the coverage model from GATK CNV by segmenting with a unified hidden Markov model, improving the statistical power to detect copy-number variation.
We validate ACNV using a purity series of the cell line HCC1143 and cancer samples from The Cancer Genome Atlas. Our results show that ACNV is able to discover regions of somatic copy-number activity accurately and with high resolution in both whole exome and whole genome sequencing data.
Citation Format: Aaron Chevalier, Lee Lichtenstein, Andrey Smirnov, Samuel K. Lee, Mehrtash Babidi, David I. Benjamin, Valentin Ruano-Rubio. GATK ACNV: allelic copy-number variation discovery from SNPs and coverage data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3581. doi:10.1158/1538-7445.AM2017-3581