Heterozygous Advantage and Segregation Distortion in Barley: Implication for Heterosis and Balancing Selection

Multiparental populations have recently become the golden standard of quantitative trait loci (QTL) mapping for plethora of traits in crop plants including maize, wheat, rice and barley. The process of developing these populations represents an evolutionary breeding experiment in which forces such as genetic drift and selection may impact the allele frequencies at the final population. Associating scenarios of segregation distortion by high-throughput genotyping for hundreds of progenies and comparing this genomic landscape with genome scan for QTL may lead to identification of signatures of selection on particular genes and pathways. More specifically, observing overlap, between significant bias to heterozygosity at certain single nucleotide polymorphic (SNP) sites and heterozygous advantage for fitness traits provides support for balancing selection. We have analyzed the high-throughput genotyping (1K SNP array) of the barley HEB-25 population (recently developed by crossing 25 wild barleys with one elite barley cultivar) for allelic segregation and for association with whole-plant phenotypic data collected under two contrasting irrigation regimes ('wet' and 'dry') in a screen house experiment during the growing season of 2014-15 at the Experimental farm of the faculty of agriculture in Rehovot.

About 150 of ~6000 SNP markers, representing ten different genomic regions were found to be significantly (p<0.05) distorted (>0.085-0.12) from the expected segregation ratio of the BC₁F₃ (0.0625). Four of these markers were found to positively and significantly (p<0.05) associate with grain yield at its heterotic state under specific irrigation regime or constitutively (3H- dry- 133cM, 5H- wet- 23cM, 6H- wet- 86cM and 7H-dry and wet- 22cM). Interestingly, from the distorted regions four were found to coincidence with major genes controlling flowering time in HEB-25 (2H-55cM-HvCEN, 4H-113Cm-vrn-H2, 5H-125cM- vrn-H1 and 7H-22cM- vrn-H3). Interestingly, under the wet regime almost all of the 150 markers exhibited heterotic effect, delaying flowering by 2-3 days, while other randomly selected genomic regions did not show such effect. These results provides support for balancing selection.