Variant analysis pipeline for accurate detection of genomic variants from transcriptome sequencing data

Abstract

The wealth of information deliverable from transcriptome sequencing (RNA-seq) is significant, however current applications for variant detection still remain a challenge due to the complexity of the transcriptome. Given the ability of RNA-seq to reveal active regions of the genome, detection of RNA-seq SNPs can prove valuable in understanding the phenotypic diversity between populations. Thus, we present a novel computational workflow named VAP (Variant Analysis Pipeline) that takes advantage of multiple RNA-seq splice aware aligners to call SNPs in non-human models using RNA-seq data only. We applied VAP to RNA-seq from a highly inbred chicken line and achieved high accuracy when compared with the matching whole genome sequencing (WGS) data. Over 65% of WGS coding variants were identified from RNA-seq. Further, our results discovered SNPs resulting from post transcriptional modifications, such as RNA editing, which may reveal potentially functional variation that would have otherwise been missed in genomic data. Even with the limitation in detecting variants in expressed regions only, our method proves to be a reliable alternative for SNP identification using RNA-seq data. The source code and user manuals are available at https://modupeore.github.io/VAP/.

Genome-Wide Development and Validation of Cost-Effective KASP Marker Assays for Genetic Dissection of Heat Stress Tolerance in Maize

“…After that, detected variants i.e., found in mapping plus SNP calling steps were filtered out to minimize false positive variant calls and considered them as priority SNPs. The priority SNPs were filtered with the set of read characteristics summarized by Adetunji et al [ 10 ] using the GATK VariantFiltration tool. In addition, after GATK VariantFiltration, VCFtools v0.1.13 [ 53 ] were used to further filter out specific variants.…”

“…The genome sequencing data obtained either experimentally from the next-generation sequencing (NGS) studies or gleaned from various databases available publicly have made it relatively simple and cheap to mine genetic variation in crop plants using various bioinformatics approaches [ 9 ]. Most of the methods detecting variation are based on sequencing data derived either from whole-genome sequencing (WGS) or whole-exome sequencing (WES) [ 10 , 11 ]. In the last few years, NGS approaches in the form of RNA-seq have been co-opted to provide global insights into the gene expression patterns to understand the genetic networks and metabolic pathways involved in maize responses to heat stress [ 12 , 13 , 14 , 15 , 16 , 17 , 18 ].…”

“…By providing an RNA signature or phenotype for a trait of interest, it allows genes and SNPs to be prioritized for marker development for a thorough understanding of that trait by genetic dissection [ 22 , 23 ]. In addition, by providing markers specifically associated with the trait of interest, it facilitates crop genetic improvement program by MAS [ 10 , 24 , 25 ].…”

RASflow: an RNA-Seq analysis workflow with Snakemake

“…It has been shown that lightweight pseudo alignment improves gene expression estimation and at the same time is computationally more efficient, compared with the standard alignment/counting methods [10]. But if the purpose of analysis is to call genomic variants, then it is still better to map the reads to the genome [11]. Considering this, a workflow should provide both quantification strategies to satisfy users with different research interests.…”

RASflow: An RNA-Seq Analysis Workflow with Snakemake

“…It has been shown that lightweight pseudo alignment improves gene expression estimation and reduces computational consumption compared with the standard alignment/counting methods [10]. But if the purpose of analysis is to call variants instead of DEA, then it is still better to map reads to genome to avoid reads mapped to spurious locations on the transcriptome [11]. Considering this, a workflow should provide both quantification strategies to satisfy users with different research interests.…”