Comparison of sequencing-based and array-based genotyping platforms for genomic prediction of maize hybrid performance

Genomic selection (GS) is a powerful tool for improving genetic gain in maize breeding. However, its routine application large-scale breeding pipelines limited by the high cost of genotyping platforms. Although sequencing-based and array-based platforms have been used GS, few studies compared prediction performance among In this study, we evaluated predictabilities four agronomic traits 305 hybrids derived from 149 parental lines subjected to sequencing (GBS), 40K SNP array, target sequence capture (TSC) using eight GS models. The GBS marker dataset yielded highest all traits, followed TSC array datasets. We investigated effect density statistical models on predictability found that 1K SNPs were sufficient achieve comparable 10K SNPs, BayesB, GBLUP, RKHS performed well, while XGBoost poorly most cases. also selected significant subsets genome-wide association study (GWAS) analyses three panels predict hybrid performance. GWAS facilitated selecting effective thus reduced cost, but depended heavily panel. conclude there still room optimization existing (GBTS) techniques integrate functional markers identified into holds great promise being an strategy developing desirable arrays.