Combining ability and analysis of genetic components of synthetic maize populations using a mixed model approach

Abstract

The aim of this study was to estimate the combined ability and analyze the genetic variance components of synthetic maize populations using diallel crossing with a mixed model approach. Twelve synthetic varieties were crossed following the complete diallel scheme without reciprocals, resulting in 66 intervarietal hybrids, which were evaluated along with the parental populations and four commercial cultivars in three different environments. An experiment was performed on 82 genotypes using a randomized complete block design with three repetitions. The evaluated traits were ear height (EH), plant height (PH) and grain yield (GY). Griffing's Model II, Method 2 and Gardner and Eberhart's Model II were combined for diallel analysis, along with the UPGMA method based on predicted breeding values (BLUP). The non-additive genetic effects were the most important for plant architecture and grain yield within each synthetic population, as based on the GCA and SCA. Clusters 1 and 2 separated by UPGMA can be included in breeding programs to obtain hybrids, and cluster 3 can be used to obtain genotypes with lower plant architectures and higher productivity in intrapopulation breeding programs.