WEAP: An automatic and accelerated pipeline for analysing multi-sample whole exome sequencing data

Abstract

Background Whole Exome Sequencing (WES) is commonly used for SNP discovery in the coding regions of the human genome and has a wide range of clinical applications. Being an intensive time-consuming task, automation is key to uncomplicating and performing straightforward data analysis. Method The WEAP workflow starts with the alignment of FASTQ files to a reference genome, variant calling, and annotation without user intervention. WEAP utilizes the GATK workflow incorporating popular NGS analysis tools such as bwa-mem2, samtools, GATK, bcftools, and anoovar coupled with GNU parallel. Results WEAP successfully identified and annotated germline and somatic variants. The major steps aligning to the reference genome, converting files, and removing duplicates in germline variant discovery were made several folds (1.5 to 3.6 folds) faster in parallel mode than in serial mode. In tumor analysis, creating a PoN from 40 samples was about 3 times faster in parallel mode. Tumor-only analysis was 1.4 to 7.7 times faster in different steps. When comparing tumor samples with matched normal tissues, the time taken was significantly reduced, making the process 1.8 to 3.6 times faster. Conclusions WEAP accepts Quality Control (QC) checked and trimmed FASTQ reads, and provides annotated variants that enable non-bioinformaticians to perform flawless variant calling from WES data. WEAP uses GNU parallel for multiple sample processing one at a time leveraging native parallel processing of the implemented tools and software to perform the analysis faster. A comparison between the parallel mode and serial mode of WEAP revealed that WEAP can be one of the best alternative tools for end-to-end analysis of WES data integrating gold standard GATK best practices workflow.