SomaticSniper: identification of somatic point mutations in whole genome sequencing data

Abstract

Motivation: The sequencing of tumors and their matched normals is frequently used to study the genetic composition of cancer. Despite this fact, there remains a dearth of available software tools designed to compare sequences in pairs of samples and identify sites that are likely to be unique to one sample.

Results: In this article, we describe the mathematical basis of our SomaticSniper software for comparing tumor and normal pairs. We estimate its sensitivity and precision, and present several common sources of error resulting in miscalls.

Availability and implementation: Binaries are freely available for download at http://gmt.genome.wustl.edu/somatic-sniper/current/, implemented in C and supported on Linux and Mac OS X.

Contact: delarson@wustl.edu; lding@wustl.edu

Supplementary information: Supplementary data are available at Bioinformatics online.

Fig. 1.

Simulations of power across various variant allele frequencies and across different depths in both samples: 30 reads (A), 60 reads (B) and 90 reads (C). Contours indicate the power for each combination of variant allele frequencies. The dashed line indicates simulations where the allele frequencies in tumor and normal are equal and thus, the contours intersecting this line indicate the expected FDR for those allele frequencies.

Fig. 2.

Association of erroneous calls with the effective 3^′ end. Kernel density estimates comparing the distribution of two different measures of variant location across three different validation results. The bandwidth (bw) for each kernel density is shown in the legend. (A) Variants that validated reference show a clear bias toward association with the effective 3^′ end. (B) All three classes of validated mutations show no clear bias in association with the end of the sequenced read.