Genetic Diversity and Selection Signatures in Synthetic-Derived Wheats and Modern Spring Wheat

Abstract

Synthetic hexaploid wheats and their derived advanced lines were subject to empirical selection in developing genetically superior cultivars. To investigate genetic diversity, patterns of nucleotide diversity, population structure, and selection signatures during wheat breeding, we tested 422 wheat accessions, including 145 synthetic-derived wheats, 128 spring wheat cultivars, and 149 advanced breeding lines from Pakistan. A total of 18,589 high-quality GBS-SNPs were identified that were distributed across the A (40%), B (49%), and D (11%) genomes. Values of population diversity parameters were estimated across chromosomes and genomes. Genome-wide average values of genetic diversity and polymorphic information content were estimated to be 0.30 and 0.25, respectively. Neighbor-joining (NJ) tree, principal component analysis (PCA), and kinship analyses revealed that synthetic-derived wheats and advanced breeding lines were genetically diverse. The 422 accessions were not separated into distinct groups by NJ analysis and confirmed using the PCA. This conclusion was validated with both relative kinship and Rogers' genetic distance analyses. EigenGWAS analysis revealed that 32 unique genome regions had undergone selection. We found that 50% of the selected regions were located in the B-genome, 29% in the D-genome, and 21% in the A-genome. Previously known functional genes or QTL were found within the selection regions associated with phenology-related traits such as vernalization, adaptability, disease resistance, and yield-related traits. The selection signatures identified in the present investigation will be useful for understanding the targets of modern wheat breeding in Pakistan.

Keywords: EigenGWAS; bread wheat; gene annotation; genetic diversity; genotyping-by-sequencing; selection signatures.

Figure 1

Distribution of minor allele frequency (A), heterozygosity (B), genetic diversity (C), and polymorphic information content (D) across 422 accessions (WP), for 145 synthetic-derived wheats (SYN-DER), 128 Pakistan cultivars (PC), and 149 advanced lines (AL) based on 18,589 SNPs.

Figure 2

The Circos plot, from outside to inside, illustrates the patterns of nucleotide diversity (A) and Tajima'D (B) values in synthetic-derived wheats (SYN-DER), commercially released Pakistan cultivars (PC), and advanced breeding lines (AL), respectively. Nucleotide diversity and Tajima'D are plotted along reference chromosomes in sliding windows of 1,000 kb with a step size of 100 kb.

Figure 3

Population structure and diversity analysis of 422 wheat accessions using 18,589 high-quality SNPs. (A) Neighbor-joining (NJ) tree, (B) principal component analysis (PCA) plot, and (C) heat map of pairwise kinship matrix with the tree shown on the top and left. The SYN-DER indicates synthetic-derived wheats; PC, commercially released Pakistan cultivars; AL, advanced breeding lines.

Figure 4

The decay of linkage disequilibrium (LD) in the whole population (WP). Pairwise LD (r²) values plotted vs. corresponding pairwise physical distance (Mb) of GBS-SNPs. The trend line of non-linear regressions against physical distance is given by the red line. The horizontal and vertical lines represent the critical value of r² (0.1) and LD decay values, respectively.

Figure 5

Miami plot showing loci under selection identified by EigenGWAS (upper for PGC and lower for F_ST) for top 10 eigenvectors (EV) based on 422 wheat accessions. PGC is the p corrected by lamdaGC in EigenGWAS. EV1-EV10 were the first 10 eigenvectors, each of which was used as phenotype for EigenGWAS analysis.

Figure 6

Pie charts depicting genomic annotation for whole-genome GBS-SNPs (18,859 SNPs) (A) and significant GBS SNPs identified by EigenGWAS (83 GBS-SNPs) (B).