The Wheat Landrace Core Collections: Potential Resource for the Breeders

Central tenet of gene bank management is to provide useful collections. Landraces possess important traits due to long-term adaptation to their native environment. As part of CIMMYT’s “Seeds of discovery-SeeD” initiative, a total of 8416 Mexican and 2403 Iranian wheat landraces were genotyped through a genotype by sequencing (GBS) approach and phenotyped under heat, drought and irrigated environments. Principal Components Analysis, PCA, was carried out to reduce the markers dimensions from 19833, and 24620 to 2000 principal component dimensions, explaining 84% and 97% of the total variance for Mexican and Iranian landrace datasets, respectively. Datasets were then reduced to matrices representing genotypic as well as phenotypic variables with the help of Gower’s distance matrix. Hierarchical Multiple Factor Analysis (HMFA) was then followed and Principal Axes (PA) was selected in such a way that the phenotype and genotype variable contribution to the total variance were 25% and 75% respectively. Coordinates for each accession on each PA used for grouping accessions and core set entries were selected through stratified-random sampling from each group using D-method.  Resulting Mexican and Iranian landrace core subsets comprised 15% of the original population. A preliminary Genome wide association analysis (GWAS) for heat-drought was performed on these core subsets and regions identified for grain yield under drought. Genomic analysis using these sets for phenology, heat, drought and diseases is underway. These collections represent a potential genetic resource for trait based analysis. The global wheat research community can efficiently utilize them for the wheat genetic improvement.