DNA extraction and library preparation

For each genotype, almost 15–20 seeds were sown in small pots along with Tift 23D2B1 in a glass house at ICRISAT, Patancheru. Leaf tissues were collected from 12–15 days old seedlings, with 5–6 seedlings per accession, each contributing approximately 100 mg of bulk leaf tissue, and stored immediately in a 96-well plate. DNA was isolated using the NucleoSpin® 96 plant II kit (Machey- Nagel, Germany), and elution of DNA was generated for library preparation. To assess the quality of genomic DNA gel electrophoresis (0.8% agarose) was performed in Tri-acetate EDTA buffer in a tank at 100 volts for 60 mins. Qualitative and quantitative checks were performed using a NanoDrop 8000 spectrophotometer, followed by the normalization of genomic DNA (10 ng/μL) was done for further GBS library preparation.

The GBS (Genotyping-By-Sequencing) method was adopted to identify genome-wide SNPs in the hybrid parental lines, as described previously [8]. First, the extracted genomic DNA was digested using ApeKI endonuclease for 2 hr at 75°C and then ligated with adapters with a unique multiplex sequence index (barcode). Next, aliquots of ligated DNA from all samples were pooled and purified to remove excess adapters. The indexed library was purified and analyzed using the Agilent Bioanalyzer 2100 (Agilent, Santa Clara, CA, USA). Subsequently, amplicons were pooled and subjected to PCR amplification. A total of 12 pM from each equimolarly pooled, index-tagged library was loaded onto eight lanes of a high-output v3 flow cell (Illumina p/n PE-401-3001-FC). The cBot-automated cluster generation system (Illumina, SY-301-20020) was employed for single-end cluster generation using the TruSeq SR Cluster Kit v3-cBot-HS (Illumina, GD-401-3001). Sequencing using single-end reads (1 × 101 bp) was carried out at Hiseq2500 platform (Illumina, SY-401-2501) and using TruSeq SBS Kit v3-HS (100 cycles) (Illumina, FC-401-3001). From HiSeq sequencing, we have obtained a total of 294,778,209 raw reads, corresponding to 1,762,042 to 5,451,371 reads per sample (average 2,704,387).