Table of Contents Author Guidelines Submit a Manuscript
BioMed Research International
Volume 2015 (2015), Article ID 456479, 11 pages
http://dx.doi.org/10.1155/2015/456479
Research Article

A Comparison of Variant Calling Pipelines Using Genome in a Bottle as a Reference

1Department of Genetics, Cell Biology and Anatomy, University of Nebraska Medical Center, Omaha, NE 68198, USA
2Bioinformatics and Systems Biology Core, University of Nebraska Medical Center, Omaha, NE 68198, USA
3Department of Biochemistry and Molecular Biology, University of Nebraska Medical Center, Omaha, NE 68198, USA
4Fred and Pamela Buffet Cancer Center, University of Nebraska Medical Center, Omaha, NE 68198, USA
5Eppley Institute for Research in Cancer and Allied Diseases, University of Nebraska Medical Center, Omaha, NE 68198, USA

Received 17 November 2014; Accepted 17 December 2014

Academic Editor: Aparup Das

Copyright © 2015 Adam Cornish and Chittibabu Guda. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

High-throughput sequencing, especially of exomes, is a popular diagnostic tool, but it is difficult to determine which tools are the best at analyzing this data. In this study, we use the NIST Genome in a Bottle results as a novel resource for validation of our exome analysis pipeline. We use six different aligners and five different variant callers to determine which pipeline, of the 30 total, performs the best on a human exome that was used to help generate the list of variants detected by the Genome in a Bottle Consortium. Of these 30 pipelines, we found that Novoalign in conjunction with GATK UnifiedGenotyper exhibited the highest sensitivity while maintaining a low number of false positives for SNVs. However, it is apparent that indels are still difficult for any pipeline to handle with none of the tools achieving an average sensitivity higher than 33% or a Positive Predictive Value (PPV) higher than 53%. Lastly, as expected, it was found that aligners can play as vital a role in variant detection as variant callers themselves.