C H A P T E R 23 Genomic selection in wheat breeding Jin Sun 1 , Maryam Khan 2 , Rabia Amir 3 , Alvina Gul 1,2 1 Department of Plant Breeding and Genetics, Cornell University, Ithaca, NY, United States; 2 Atta-ur-Rahman School of Applied Biosciences (ASAB), National University of Sciences and Technology (NUST), Islamabad, Pakistan; 3 Department of Plant Biotechnology, Atta-ur-Rahman School of Applied Biosciences (ASAB), National University of Sciences and Technology (NUST), Islamabad, Pakistan OUTLINE 1. Introduction 321 2. Approaches to improve genomic selection accuracy in wheat 322 2.1 Genomic selection models 323 2.2 Genotypeenvironment interactions 323 2.3 Platforms for high-throughput phenotyping 326 3. Application of genomic selection in wheat 326 3.1 Wheat grain yield and genomic selection 327 3.2 Wheat disease resistance and genomic selection 327 3.3 Genomic selection and other traits 328 4. Summary 328 References 328 1. Introduction Genomic selection (GS) is a model-based approach in plant breeding that utilizes genomic estimated breeding values (GEBVs) of breeding lines to predict breeding outcomes in an effective and efficient manner. It basically es- tablishes links between phenotypes and genetic markers to accelerate the genetic gain in plant breeding (Wang et al., 2018). Initially, it was established in the animal breeding because of inability of animals to replicate and the high cost of phenotyping (Rutkoski et al., 2017). GS gained limelight in plant breeding because it achieves more and compre- hensive selection as compared with other conventional breeding tools that mostly rely on phenotype selection. It uses genomic prediction models based on genome-wide prediction markers and phenotypic traits of the training population (TRN) to predict the GEBVs of the testing population (TSN) that only have genotypic data. Then the GEBVs of those lines in TSN will be utilized to process the selection for the next breeding cycle as described in Fig. 23.1 (Lorenz et al., 2011). However, the advancement of GS in plant breeding field is relatively far away behind as compared with animal breeding, in which the initial implantation was performed in 2007 based on the simulated data in maize (Bernardo and Yu, 2007). GS is a sensitive selection process that even accounts small-effect markers with potential to interfere with signif- icance of test. It has showed significant advantages in plant breeding as compared with traditional marker-assisted selection (MAS) especially for complex quantitative traits with low heritability and regulated by various loci with small effects. It is able to capture more variations and to improve selections that involve genome-wide spread genetic markers. Moreover, GS reduces breeding cycles by improving the genetic gain per unit time, which is demonstrated by the breeder’s equation (G ¼ irs A Y , where G denotes gain per year, i denotes selection intensity, r denotes selection accuracy, s A denotes square root of narrow sense heritability, and Y denotes time in years for a cycle of selection). According to the equation, larger genetic gain can be achieved, compared with traditional phenotypic selection (PS), 321 Climate Change and Food Security with Emphasis on Wheat Copyright © 2020 Elsevier Inc. All rights reserved. https://doi.org/10.1016/B978-0-12-819527-7.00023-6