23
Computational resources to facilitate variant discovery and analysis
Next generation sequencing has facilitated the sequencing of large numbers of individuals for variant detection. A challenge facing the livestock industry is the establishment of efficient workflows to process raw sequence data to accurate variant data such that implementation of whole genome sequence information in livestock breeding programs can be accomplished. Within the iPlant infrastructure, we have implemented currently available variant calling techniques and have applied them to a dairy cattle data set. The Burrows-Wheeler aligner (BWA) was used to align paired end Illumina reads from 66 bulls to the bos taurus UMD3.1 reference assembly. A pipeline, integrating data preparation, insertion and deletion realignment, base quality score recalibration and variant calling, was created to manage variant calling comparisons. Variant calling was done using three different calling methods: UnifiedGenotyper of the Genome Analysis Tool Kit, SAMTools Mpileup and Platypus. A total of nine different options were compared with these methods. Significant variation in the ability of the different methods calling of variants was observed. Furthermore, tremendous variation in the concordance of variant calling and SNPchip genotyping were observed as well. These results indicate that a more detailed analysis and/or the use of a combination of callers is necessary if the livestock industry is to fully utilize genome re-sequencing information to improve livestock breeding.
The authors gratefully acknowledge financial support from the Swiss Cattle Breeders Association (ASR) and the Swiss Commission for Technology and Innovation (CTI), and USDA-NIFA project 2013-01001.
Keywords: next generation sequencing