SNP-SVant: A Computational Workflow to Predict and Annotate Genomic Variants in Organisms Lacking Benchmarked...

Area of Science:

• Genomics
• Bioinformatics

Background:

• Whole-genome sequencing is crucial for understanding population genomic variation.
• Existing variant calling tools often require benchmarked datasets to distinguish sequencing errors from true variants.
• This limitation hinders variant prediction in organisms lacking such resources.

Purpose of the Study:

• To develop an integrated, flexible, and computationally efficient bioinformatic workflow called SNP-SVant.
• To enable high-confidence prediction of single nucleotide polymorphisms (SNPs) and structural variations (SVs).
• To address the challenge of variant calling in organisms without benchmarked variant datasets.

Main Methods:

• SNP-SVant integrates variant calling using Genome Analysis ToolKit (GATK) for SNPs and Genome Rearrangement IDentification Software Suite (GRIDSS) for SVs.
• It employs multiple rounds of statistical recalibration to enhance variant prediction precision in the absence of benchmarked data.
• The workflow utilizes a workflow management system for scalability and efficient resource utilization, with checkpoint steps to minimize redundant computations.

Main Results:

• SNP-SVant successfully predicts high-confidence SNPs and SVs, including small insertions and deletions.
• The workflow offers user-configurable options to balance accuracy and sensitivity.
• It provides variant quality assessment metrics and format conversion capabilities (VCF to aligned FASTA) for downstream analyses.

Conclusions:

• SNP-SVant advances variant prediction capabilities, particularly for organisms lacking benchmarked data.
• Its integrated approach and scalability make it a valuable tool for population genomics.
• This workflow enhances the ability to associate genotypes with phenotypes by providing a comprehensive view of genomic alterations.