Genotyping melon accessions using GBS

Genotyping of the melon accessions was performed following the GBS protocol described in Elshire et al. (2011 (link)), using the ApeKI restriction enzyme (NEB, Beverly, MA). GBS libraries were sequenced on a HiSeq 2500 system with read length of 101 bp. The TASSEL 5.0 GBS discovery pipeline (Glaubitz et al. 2014 (link)) was used for SNP calling. Briefly, raw GBS sequencing reads that possessed a barcode and a restriction enzyme cut site were identified using GBSSeqToTagDBPlugin in TASSEL with parameters ‘-kmerLength 90 -minKmerL 30 -mnQS 10 -c 10 -maKmerNum 200000000’. Tags were retrieved and reformatted using TagExportToFastqPlugin, and those supported by at least ten reads were kept and then mapped to the melon reference genome sequence (v3.5.1) (Garcia-Mas et al. 2012 ) using BWA v0.7.13 (Li and Durbin 2009 (link)) with default parameters. Based on the alignments, positions of aligned tags were determined using SAMtoGBSdbPlugin, and SNPs were identified from the aligned tags using DiscoverySNPCallerPluginV2 with default parameters. SNPs were filtered based on their missing data rate and minor allele frequencies (MAF) using VCFtools v 0.1.15 (Danecek et al. 2011 (link)).