Genetic architecture of agronomic and quality traits in a nested association mapping population of spring wheat

Abstract

Abstract Germplasm collections are rich sources of genetic variation to improve crops for many valuable traits. Nested association mapping (NAM) populations can overcome the limitations of genome‐wide association studies (GWAS) in germplasm collections by reducing the effect of population structure. We exploited the genetic diversity of the USDA‐ARS wheat (Triticum aestivum L.) core collection by developing the Spring Wheat Multiparent Introgression Population (SWMIP). To develop this population, twenty‐five core parents were crossed and backcrossed to the Minnesota spring wheat cultivar RB07. The NAM population and 26 founder parents were genotyped using genotyping‐by‐sequencing and phenotyped for heading date, height, test weight, and grain protein content. After quality control, 20,312 markers with physical map positions were generated for 2,038 recombinant inbred lines (RILs). The number of RILs in each family varied between 58 and 96. Three GWAS models were utilized for quantitative trait loci (QTL) detection and accounted for known family stratification, genetic kinship, and both covariates. GWAS was performed on the whole population and also by bootstrap sampling of an equal number of RILs from each family. Greater power of QTL detection was achieved by treating families equally through bootstrapping. In total 16, 15, 12, and 13 marker‐trait associations (MTAs) were identified for heading date, height, test weight, and grain protein content, respectively. Some of these MTAs were coincident with major genes known to control the traits, but others were novel and contributed by the wheat core parents. The SWMIP will be a valuable source of genetic variation for spring wheat breeding.