Genomic prediction of drought tolerance during seedling stage in maize using low-cost molecular markers

Abstract

AbstractDrought tolerance in maize is a complex and polygenic trait, especially in the seedling stage. In plant breeding, complex genetic traits can be improved by genomic selection (GS), which has become a practical and effective breeding tool. In the present study, a natural maize population named Northeast China core population (NCCP) consisting of 379 inbred lines were genotyped with diversity arrays technology (DArT) and genotyping-by-sequencing (GBS) platforms. Target traits of seedling emergence rate (ER), seedling plant height (SPH), and grain yield (GY) were evaluated under two natural drought stress environments in northeast China. Adequate genetic variations were observed for all the target traits, but they were divergent across environments. Similarly, the heritability of the target trait also varied across years and environments, the heritabilities in 2019 (0.88, 0.82, 0.85 for ER, SPH, GY) were higher than those in 2020 (0.65, 0.53, 0.33) and cross-2-years (0.32, 0.26, 0.33). In total, three marker datasets, 11,865 SilicoDArT markers obtained from the DArT-seq platform, 7837 SNPs obtained from the DArT-seq platform, and 91,003 SNPs obtained from the GBS platform, were used for GS analysis after quality control. The results of phylogenetic trees showed that broad genetic diversity existed in the NCCP population. Genomic prediction results showed that the average prediction accuracies estimated using the DArT SNP dataset under the two-fold cross-validation scheme were 0.27, 0.19, and 0.33, for ER, SPH, and GY, respectively. The result of SilicoDArT is close to the SNPs from DArT-seq, those were 0.26, 0.22, and 0.33. For the trait with lower heritability, the prediction accuracy can be improved using the dataset filtered by linkage disequilibrium. For the same trait, the prediction accuracies estimated with two DArT marker datasets were consistently higher than that estimated with the GBS SNP dataset under the same genotyping cost. The prediction accuracy was improved by controlling population structure and marker quality, even though the marker density was reduced. The prediction accuracies were improved by more than 30% using the significant-associated SNPs. Due to the complexity of drought tolerance under the natural stress environments, multiple years of data need to be accumulated to improve prediction accuracy by reducing genotype-by-environment interaction. Modeling genotype-by-environment interaction into genomic prediction needs to be further developed for improving drought tolerance in maize. The results obtained from the present study provides valuable pathway for improving drought tolerance in maize using GS