A Pipeline for Markers Selection Using Restriction Site Associated DNA Sequencing (RADSeq)

Motivation: The discovery and assessment genetic variants for Next Generation Sequencing (NGS), including Restriction site Associated DNA sequencing (RADSeq), is an important task in bioinformatics and comparative genetics. The genetic variants can be single-nucleotide polymorphisms (SNPs), insertions and deletions (Indels) when compared to a reference genome. Usually, the short reads are aligned to a reference genome at first using NGS alignment software, such as the Burrows- Wheeler Aligner (BWA). The alignment is usually stored into a BAM file, a binary format of standard SAM (Sequence Alignment/Map) protocol. Then analysis software, such as Genome analysis Toolkit (GATK) or SAMTools [30] [31], together with scripts written in R programming language, could provide an efficient solution for calling variants. We focused on RADSeq-based marker selection for Arabidopsis thaliana. RADSeq consists short reads that do not cover the whole reference genome. Finally, SNPs as output in Variant Call Format (VCF) have been visualized by Integrative Genomics Viewer (IGV) software. We found that the visualization of SNPs and Indels is helpful and provides us with valuable insights on marker selection. We found that applying Chi-Square test for all target genotypes, which are homozygous reference 0/0, heterozygous variants 0/1 and homozygous variants 1/1, to test Hardy-Weinberg Equilibrium (HWE) in order to reduce false positive rate significantly and we showed that our pipeline is efficient in RADSeq-based marker selection