Genome-wide generation and genotyping of informative SNPs to scan molecular signatures for seed yield in chickpea

Abstract

We discovered 2150 desi and 2199 kabuli accessions-derived SNPs by cultivar-wise individual assembling of sequence-reads generated through genotyping-by-sequencing of 92 chickpea accessions. Subsequent large-scale validation and genotyping of these SNPs discovered 619 desi accessions-derived (DAD) SNPs, 531 kabuli accessions-derived (KAD) SNPs, 884 multiple accessions-derived (MAD) SNPs and 1083 two accessions (desi ICC 4958 and kabuli CDC Frontier)-derived (TAD) SNPs that were mapped on eight chromosomes. These informative SNPs were annotated in coding/non-coding regulatory sequence components of genes. The MAD-SNPs were efficient to detect high intra-specific polymorphic potential and wide natural allelic diversity level including high-resolution admixed-population genetic structure and precise phylogenetic relationship among 291 desi and kabuli accessions. This signifies their effectiveness in introgression breeding and varietal improvement studies targeting useful agronomic traits of chickpea. Six trait-associated genes with SNPs including quantitative trait nucleotides (QTNs) in combination explained 27.5% phenotypic variation for seed yield per plant (SYP). A pentatricopeptide repeat (PPR) gene with a synonymous-coding SNP/QTN significantly associated with SYP trait was found most-promising in chickpea. The essential information delineated can be of immense utility in genomics-assisted breeding applications to develop high-yielding chickpea cultivars.

Figure 1

Strategy adopted in the present study to generate high-quality genotyping data of diverse kinds of informative SNPs (DAD-, KAD-, MAD- and TAD-SNPs) from 291 desi and kabuli accessions at a genome-wide scale to be deployed for multiple genomics-assisted breeding applications in chickpea.

Figure 2

Proportionate distribution of MAD- and TAD-SNPs in different coding (synonymous and non-synonymous) and non-coding (intron, URR and DRR) sequence components of genes and intergenic regions annotated from desi chickpea genome. The gene annotation of desi genome was considered as reference to infer the coding DNA sequence (CDS)/exons, introns, up/downstream regulatory region (URR/DRR) sequence components of genes. Digits within the round and square Parentheses indicate the (number of gene-derived SNPs) and [number of genes with SNPs], respectively, representing each class of coding and non-coding regions of chickpea genes.

Figure 3

Unrooted phylogenetic tree and population structure (with population number K = 2) illustrating the evolutionary relationships and best possible population genetic structure, respectively, among 291 desi and kabuli chickpea accessions using 619 DAD (A) 531 KAD (B) 884 MAD (C) and 1083 TAD (D) SNPs. The mapped markers assigned accessions into two populations (POP I and POP II) which essentially corresponded to their geographical as well as cultivar-specific origination. Red and green colours correspond to POP I and POP II, respectively, as defined by both phylogenetic tree and population genetic structure. The accessions represented by vertical bars along the horizontal axis were classified into K colour segments as per their estimated membership fraction in each K cluster.

Figure 4

Frequency distribution of seed yield per plant (SYP) trait variation estimated among 291 desi and kabuli chickpea germplasm accessions (association panel) field phenotyped at multiple geographical locations/years.

Figure 5

GWAS-derived Manhattan plot depicting significant P-values associated with seed yield per plant (SYP) using 619 DAD-, 531 KAD-, 884 MAD- and 1083 TAD-SNPs in chickpea. The genomic distribution of SNPs physically mapped on eight chromosomes of chickpea genome are designated by the x-axis. The y-axis specifies the −log₁₀ (P)-value for significant association of SNP loci with SYP trait. The SNPs significantly associated with SYP trait at a cut-off P value ≤ 10⁻⁶ are highlighted with dotted lines.