Genome-Wide Association Studies of Disease and Drought-Relative Traits using Whole Genome Re-Sequencing Data in Chickpea

Next-generation sequencing (NGS) technology offers a cheap and high-throughput genotyping option to discover genome variation and selection signatures in less utilised crop species, such as chickpea.

We performed whole genome re-sequencing (WGRS) of 64 Australian chickpea varieties, four Indian landraces, and one wild chickpea species (Cicer reticulatum) with 5-15X coverage. Alignment of 1.2 billion Illumina paired-end reads to the draft Kabuli genome sequence of chickpea resulted in the identification of over 800,000 SNPs. To handle the high error rate of NGS data, allele frequencies were estimated using site frequency spectrum as prior leading to improved inference of population genetic parameters. Population structure analysis reveals distinct groups among varieties and narrow genetic diversity in recently released varieties. Several regions of the chickpea genome are under positive selection based on Tajima’s D test. Both Fst genome scan and genome-wide association studies (GWAS) identify a 100kb region on chickpea chromosome 4 that is significantly associated with ascochyta blight resistance, a fungus disease that severely impacts the chickpea production in Australia and other regions of the world. This region is co-located in a large QTL interval of 7Mb~30Mb confirmed previously by three different mapping populations genotyped at low density with SSR or SNP markers. This 100kb region has been validated by GWAS of another 132 advance lines with ~140,000 SNPs. In total, 13 predicted genes are located in this region including NBS-LRR receptor-like kinase, wall-associated kinase, zinc finger protein and serine/threonine protein kinase. One significant SNP located in the coding sequence of a predicted gene leads to amino acid substitution. This demonstrates the power of combining WGRS data with relatively simple traits in fast developing “functional makers” for marker-assisted selection.

The 132 advance lines were phenotyped with 12 yield and yield relative traits in three drought-prone field environments. GWAS analyses identify 35 SNPs significantly (p< 3.45e-07) associated with 6 traits in total. The result of genomic prediction using these lines as training population will be presented as well.