Email updates

Keep up to date with the latest news and content from BMC Research Notes and BioMed Central.

Open Access Highly Accessed Technical Note

RVD: a command-line program for ultrasensitive rare single nucleotide variant detection using targeted next-generation DNA resequencing

Anna Cushing1, Patrick Flaherty24, Erik Hopmans3, John M Bell3 and Hanlee P Ji13*

Author Affiliations

1 Division of Oncology, Department of Medicine, Stanford University School of Medicine, Stanford, CA 94305, USA

2 2Department of Biochemistry, Stanford University School of Medicine, Stanford, CA 94305, USA

3 Stanford Genome Technology Center, Stanford University, Palo Alto, CA 94304, USA

4 Department of Biomedical Engineering, Worcester Polytechnic Institute, Worcester, MA 01605, USA

For all author emails, please log on.

BMC Research Notes 2013, 6:206  doi:10.1186/1756-0500-6-206

Published: 23 May 2013

Abstract

Background

Rare single nucleotide variants play an important role in genetic diversity and heterogeneity of specific human disease. For example, an individual clinical sample can harbor rare mutations at minor frequencies. Genetic diversity within an individual clinical sample is oftentimes reflected in rare mutations. Therefore, detecting rare variants prior to treatment may prove to be a useful predictor for therapeutic response. Current rare variant detection algorithms using next generation DNA sequencing are limited by inherent sequencing error rate and platform availability.

Findings

Here we describe an optimized implementation of a rare variant detection algorithm called RVD for use in targeted gene resequencing. RVD is available both as a command-line program and for use in MATLAB and estimates context-specific error using a beta-binomial model to call variants with minor allele frequency (MAF) as low as 0.1%. We show that RVD accepts standard BAM formatted sequence files. We tested RVD analysis on multiple Illumina sequencing platforms, among the most widely used DNA sequencing platforms.

Conclusions

RVD meets a growing need for highly sensitive and specific tools for variant detection. To demonstrate the usefulness of RVD, we carried out a thorough analysis of the software’s performance on synthetic and clinical virus samples sequenced on both an Illumina GAIIx and a MiSeq. We expect RVD can improve understanding the genetics and treatment of common viral diseases including influenza. RVD is available at the following URL:http://dna-discovery.stanford.edu/software/rvd/ webcite.

Keywords:
Next-generation sequencing; Rare variant detection; Genotyping