Opossum – pre-processing sequencing data for reliable SNP variant detection

Posted by: RNA-Seq Blog in Publications, SNP Detection February 28, 2017 6,917 Views

Identifying variants from RNA-seq (transcriptome sequencing) data is a cost-effective and versatile alternative to whole-genome sequencing. However, current variant callers do not generally behave well with RNA-seq data due to reads encompassing intronic regions. Researchers from the Wellcome Trust Centre for Human Genetics and the Centre for Evolution and Cancer have developed a software programme called Opossum to address this problem. Opossum pre-processes RNA-seq reads prior to variant calling, and although it has been designed to work specifically with Platypus, it can be used equally well with other variant callers such as GATK HaplotypeCaller. In this work, the researchers show that using Opossum in conjunction with either Platypus or GATK HaplotypeCaller maintains precision and improves the sensitivity for SNP detection compared to the GATK Best Practices pipeline. In addition, using it in combination with Platypus offers a substantial reduction in run times compared to the GATK pipeline so it is ideal when there are only limited time or computational resources available.

Precision as a function of the number of supporting bases

RNA-seq data mapped with TopHat2 v2.0.12. GATK HC stands for GATK HaplotypeCaller v3.4

Availability – the source code is avialble at: https://github.com/BSGOxford/Opossum

Oikkonen L, Lise S. (2017) Making the most of RNA-seq: Pre-processing sequencing data with Opossum for reliable SNP variant detection. Wellcome Open Res 2:6. [article]