Genome-wide discovery of DNA polymorphisms among chickpea cultivars with contrasting seed size/weight and their functional relevance

Abstract

Seed size/weight is a major agronomic trait which determine crop productivity in legumes. To understand the genetic basis of seed size determination, we sought to identify DNA polymorphisms between two small (Himchana 1 and Pusa 362) and two large-seeded (JGK 3 and PG 0515) chickpea cultivars via whole genome resequencing. We identified a total of 75535 single nucleotide polymorphisms (SNPs), 6486 insertions and deletions (InDels), 1938 multi-nucleotide polymorphisms (MNPs) and 5025 complex variants between the two small and two large-seeded chickpea cultivars. Our analysis revealed 814, 244 and 72 seed-specific genes harboring DNA polymorphisms in promoter or non-synonymous and large-effect DNA polymorphisms, respectively. Gene ontology analysis revealed enrichment of cell growth and division related terms in these genes. Among them, at least 22 genes associated with quantitative trait loci, and those involved in cell growth and division and encoding transcription factors harbored promoter and/or large-effect/non-synonymous DNA polymorphisms. These also showed higher expression at late-embryogenesis and/or mid-maturation stages of seed development in the large-seeded cultivar, suggesting their role in seed size/weight determination in chickpea. Altogether, this study provided a valuable resource for large-scale genotyping applications and a few putative candidate genes that might play crucial role in governing seed size/weight in chickpea.

Discovery of DNA polymorphisms via whole genome resequencing and their functional relevance in salinity stress response in chickpea

“…The whole genome sequencing library for each sample was prepared and sequencing was performed using the Illumina Hiseq-2000 platform (Illumina) to obtain 100-nt long paired-end reads as described earlier (Rajkumar et al, 2021). The raw data were preprocessed to filter-out low-quality reads and the reads containing adaptor sequences using NGS QC Toolkit (v2.3) at default parameters (Patel & Jain, 2012).…”

“…The raw data were preprocessed to filter-out low-quality reads and the reads containing adaptor sequences using NGS QC Toolkit (v2.3) at default parameters (Patel & Jain, 2012). Only the filtered high-quality reads were mapped to the kabuli chickpea reference genome (v1.0; Varshney et al, 2013) using the BWA software (v 6.2) at default parameters as described earlier (Rajkumar et al, 2021).…”

“…DNA polymorphisms in the six chickpea genotypes as compared to the reference chickpea genome were identified with criteria of mapping-quality ≥ 30, coverage ≥ 10, and alternate-fraction ≥ 0.9 using the FreeBayes (v 0.8.7) software as described earlier (Rajkumar et al, 2018(Rajkumar et al, , 2021. Furthermore, we identified a high-confidence set of DNA polymorphisms that differentiated salinity-sensitive and salinity-tolerant chickpea genotypes using the following criteria.…”

“…To analyze the influence of non-synonymous SNPs on the structural integrity of the encoded proteins, we determined the best PDB protein structure template for each selected protein via the Phyre2 Investigator (Kelley et al, 2016) in intensive mode as described earlier (Rajkumar et al, 2021). The three-dimensional (3D) structure of the proteins was determined using the best PDB match of the selected proteins.…”

“…The advancement of high-throughput sequencing technologies along with the availability of reference genome sequences of chickpea (Jain et al, 2013;Varshney et al, 2013) have provided excellent opportunities to perform large-scale genotyping at affordable costs. Whole genome resequencing (WGR) has been performed in numerous chickpea genotypes to examine diversity, domestication, and their plausible influence on agronomic traits (Li et al, 2018;Rajkumar et al, 2018Rajkumar et al, , 2021Thudi et al, 2016;Varshney et al, 2019). However, molecular genetic analysis of salinity stress response in chickpea has been limited largely to comparatively low-throughput technologies (Pushpavalli et al, 2015;Soren et al, 2020;Vadez et al, 2012).…”