Vol.:(0123456789) 1 3 Theoretical and Applied Genetics https://doi.org/10.1007/s00122-017-3037-7 ORIGINAL ARTICLE Genetic analysis of multi‑environmental spring wheat trials identifes genomic regions for locus‑specifc trade‑ofs for grain weight and grain number Sivakumar Sukumaran 1  · Marta Lopes 2  · Susanne Dreisigacker 1  · Matthew Reynolds 1 Received: 11 September 2017 / Accepted: 1 December 2017 © Springer-Verlag GmbH Germany, part of Springer Nature 2017 Abstract Key message GWAS on multi-environment data identified genomic regions associated with trade-offs for grain weight and grain number. Abstract Grain yield (GY) can be dissected into its components thousand grain weight (TGW) and grain number (GN), but little has been achieved in assessing the trade-of between them in spring wheat. In the present study, the Wheat Association Mapping Initiative (WAMI) panel of 287 elite spring bread wheat lines was phenotyped for GY, GN, and TGW in ten envi- ronments across diferent wheat growing regions in Mexico, South Asia, and North Africa. The panel genotyped with the 90 K Illumina Infnitum SNP array resulted in 26,814 SNPs for genome-wide association study (GWAS). Statistical analysis of the multi-environmental data for GY, GN, and TGW observed repeatability estimates of 0.76, 0.62, and 0.95, respectively. GWAS on BLUPs of combined environment analysis identifed 38 loci associated with the traits. Among them four loci—6A (85 cM), 5A (98 cM), 3B (99 cM), and 2B (96 cM)—were associated with multiple traits. The study identifed two loci that showed positive association between GY and TGW, with allelic substitution efects of 4% (GY) and 1.7% (TGW) for 6A locus and 0.2% (GY) and 7.2% (TGW) for 2B locus. The locus in chromosome 6A (79–85 cM) harbored a gene TaGW2-6A. We also identifed that a combination of markers associated with GY, TGW, and GN together explained higher variation for GY (32%), than the markers associated with GY alone (27%). The marker-trait associations from the present study can be used for marker-assisted selection (MAS) and to discover the underlying genes for these traits in spring wheat. Abbreviations WAMI The wheat association Mapping Initiative BLUPs Best linear unbiased predictions MLM Mixed linear models GLM Generalized linear models Introduction Wheat (Triticum aestivum L.) provides 20% of the total calo- ries and 20% of plant-derived protein to the world popu- lation (Food and Agricultural Organization of the United Nations, 2010). However, the production levels need to be increased by 70% to meet the projected food requirements by 2050 (Ray et al. 2012). Even though progress has been made through conventional breeding approaches in increas- ing genetic gains of grain yield (GY) of spring bread wheat, it is less than 1% per year (Sharma et al. 2012; Aisawi et al. 2015; Crespo-Herrera et al. 2017). To meet the predicted demand, it is important to complement the conventional approaches through molecular breeding for complex traits (Reynolds and Langridge 2016). Even though GY is the most important trait in any plant breeding program, there still exists a large gap in understanding the genetic and molecular mechanism of the trait and its components (Val- luru et al. 2014). Linkage mapping and genome-wide association stud- ies (GWAS) are two methods widely used to identify and Communicated by Mark E. Sorrells. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00122-017-3037-7) contains supplementary material, which is available to authorized users. * Sivakumar Sukumaran s.sukumaran@cgiar.org 1 Global Wheat Program, International Maize and Wheat Improvement Center (CIMMYT), Apdo. Postal 6-641, Mexico City 06600, Mexico 2 CIMMYT, P.O. Box 39, Emek, Ankara 06511, Turkey