Premium
Improving Genomic Prediction for Pre‐Harvest Sprouting Tolerance in Wheat by Weighting Large‐Effect Quantitative Trait Loci
Author(s) -
Moore Jessica K.,
Manmathan Harish K.,
Anderson Victoria A.,
Poland Jesse A.,
Morris Craig F.,
Haley Scott D.
Publication year - 2017
Publication title -
crop science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.76
H-Index - 147
eISSN - 1435-0653
pISSN - 0011-183X
DOI - 10.2135/cropsci2016.06.0453
Subject(s) - biology , quantitative trait locus , germplasm , single nucleotide polymorphism , trait , preharvest , marker assisted selection , snp , genotyping , selection (genetic algorithm) , genetic marker , genetics , microbiology and biotechnology , genotype , agronomy , botany , gene , machine learning , postharvest , computer science , programming language
Preharvest sprouting (PHS) is a major problem in wheat ( Triticum aestivum L.) that occurs when grains in a mature spike germinate before harvest, resulting in reduced yield, quality, and grain sale price. Improving PHS tolerance is a challenge to wheat breeders because it is quantitatively inherited and tedious to score. Genomic selection (GS) is particularly useful for predicting phenotypes that are costly and time consuming to assess. In our study, single nucleotide polymorphism (SNP) markers obtained by genotyping‐by‐sequencing were used to identify significant marker trait associations and develop predictive models for PHS tolerance. A panel of 1118 breeding lines and cultivars (genotypes) representative of U.S. Great Plains hard winter wheat germplasm was scored for PHS tolerance over multiple years. A genomewide association approach was used to identify quantitative trait loci (QTL) among the individuals. Two primary factors were examined for their influence on model accuracy: the effect of including identified QTL and kernel color as fixed effects in the model and increasing marker number. Model accuracy did not improve with kernel color information, but weighting QTL increased predictive performance. Thus, the combination of marker‐assisted and genomic selection outperformed all other methods. Optimum marker number was reached at 4000 SNPs. Overall, model accuracies were promising (0.49 to 0.62) and confirm effectiveness of GS for predicting PHS tolerance in wheat.