Population structure and linkage disequilibrium in elite barley breeding germplasm from the United States

Abstract

Cultivated barley is known to have a complex population structure and extensive linkage disequilibrium (LD). To conduct robust association mapping (AM) studies of economically important traits in US barley breeding germplasm, population structure and LD decay were examined in a complete panel of US barley breeding germplasm (3840 lines) genotyped with 3072 single nucleotide polymorphisms (SNPs). Nine subpopulations (sp1‒sp9) were identified by the program STRUCTURE and subsequently confirmed by principle component analysis (PCA). Out of the nine subpopulations, seven were very similar to the respective subpopulations identified by Hamblin et al. (2010) which were based on half of the germplasm and half of the SNP markers, but two subpopulations were found to be new. One subpopulation was dominated by six-rowed spring lines from Utah State University (UT) and the other was composed of six-rowed spring lines from multiple breeding programs (USDA-ARS Aberdeen (AB), Busch Agricultural Resources Inc. (BA), UT, and Washington State University (WA)). LD was found to decay across a range from 4.0 to 19.8 cM. This result indicates that the germplasm genotyped with 3072 SNPs would be robust for mapping and possibly identifying the causal polymorphisms contributing to disease resistance and perhaps other traits.

Fig. 1

Boxplots of minor allele frequency (MAF) in US barley breeding germplasm Five statistics are represented as bars in each boxplot from bottom to top: the smallest observation, lower quartile, median, upper quartile, and largest observation, respectively. Data points which lie outside this range are extreme values. Heterozygosity rate is given at the bottom under each breeding program. Breeding programs: University of Minnesota (MN), North Dakota State University (N6), USDA-ARS Aberdeen (AB), Utah State University (UT), Busch Agricultural Resources Inc. (BA), North Dakota State University (N2), Washington State University (WA), Montana State University (MT), Oregon State University (OR), Virginia Polytechnic Institute and State University (VT)

Fig. 2

Three-dimensional plot of the complete panel (CAP) of US barley breeding germplasm based on PCA The first PC was due predominantly (20.3%) to different row types (i.e., six-rowed vs. two-rowed), and the second PC (10.5%) was due to winter/facultative vs. spring types. PC3 only explained 3.9% of the total variation, but it separated the two subgroups within the two-rowed spring lines. The nine subpopulations (sp1–sp9) defined by STRUCTURE were color-coded and numbered. The three major clusters defined by PCA also were labeled. Breeding programs: University of Minnesota (MN), North Dakota State University (N6), USDA-ARS Aberdeen (AB), Utah State University (UT), Busch Agricultural Resources Inc. (BA), North Dakota State University (N2), Washington State University (WA), Montana State University (MT), Oregon State University (OR), Virginia Polytechnic Institute and State University (VT)