Choice of reference-guided sequence assembler and SNP caller for analysis of Listeria monocytogenes short-read sequence data greatly influences rates of error.

Arthur W. Pightling,Franco Pagotto,Nicholas Petronella

doi:10.1186/s13104-015-1689-4

Abstract

BackgroundThe influences that different programs and conditions have on error rates of single-nucleotide polymorphism (SNP) analyses are poorly understood. Using Illumina short-read sequence data generated from Listeria monocytogenes strain HPB5622, we assessed the performance of four SNP callers (BCFtools, FreeBayes, UnifiedGenotyper, VarScan) under a variety of conditions, including: (1) a range of sequencing coverages; (2) use of four popular reference-guided assemblers (Burrows-Wheeler Aligner, Novoalign, MOSAIK, SMALT); (3) with and without read quality trimming and filtering; and (4) use of different reference sequences.ResultsAt 8-fold coverage the proportions of true positive calls ranged from 0.22 to 25.00 % when reads were aligned to a nearly identical reference (0.000096 % distant). Calls made when reads were aligned to a non-identical reference (0.85 % distant) were from 92.54 to 98.88 % accurate. At 79-fold coverage accuracies ranged from 3.95 to 20.00 % with the nearly identical reference and 93.80–98.75 % with the non-identical reference. Read preprocessing significantly changed the numbers of false positive calls made, from a 65.24 % decrease to a 54.55 % increase.ConclusionsThe combinations of reference-guided sequence assemblers and SNP callers greatly influenced not only the numbers of true and false positive sites but also the proportions of true positive calls relative to the total numbers of calls made. Furthermore, the efficacy of different assembler and caller combinations changed dramatically with the different conditions tested. Researchers should consider whether identifying the greatest numbers of true positive sites, reducing the numbers of false positive calls, or achieving the highest accuracies are desired.Electronic supplementary materialThe online version of this article (doi:10.1186/s13104-015-1689-4) contains supplementary material, which is available to authorized users.

Highlights

The influences that different programs and conditions have on error rates of single-nucleotide polymorphism (SNP) analyses are poorly understood
We assessed the ability of four commonly used singlenucleotide polymorphism (SNP) callers (BCFtools, FreeBayes, UnifiedGenotyper, and VarScan) to identify SNPs from alignments of eight sets of Listeria monocytogenes strain HPB5622 genomic DNA sequence data of varying quality generated on an Illumina MiSeq benchtop sequencer
Genome sequence assembly is a formidable computational challenge that can be influenced by several factors, including: (1) the amount sequence coverage, (2) the algorithm used by the reference-guided sequence assemblers to place each read, (3) the distance of the reference sequence from the subject, and (4) whether read quality trimming and filtering has been performed

Summary

Introduction

The influences that different programs and conditions have on error rates of single-nucleotide polymorphism (SNP) analyses are poorly understood. Low sequencing coverage [16] and the use of genetically distant reference sequences [17] provide additional computational challenges for both reference-guided sequence assemblers and SNP callers [18], especially around regions of repeated DNA sequence [19]. These issues may result in diminished detection of true SNP differences (true positive calls) and increased numbers of misidentified SNPs (false positive calls). Several factors may influence the accurate identification of true nucleotide differences: (1) sequencing coverage, (2) read preprocessing, (3) availability of an appropriate reference sequence, (4) selection of short-read sequence assembler, and (5) one’s choice of SNP calling software

Methods

Results

Conclusion