Heterotic prediction of hybrid performance based on genome-wide SNP markers and the phenotype of parental inbred lines in heading Chinese cabbage (Brassica rapa L. ssp. pekinensis)

Chinese cabbage (Brassica rapa L. ssp. Pekinensis) is the most widely consumed leafy vegetable in Asia. Most commercial cultivars are F1 hybrids with high yield and good quality. To increase the breeding efficiency, heterotic prediction of hybrid performance was studied. Fourteen parental lines of Chinese cabbage were studied in a half-diallel cross to produce 91 hybrids. The average mid-parent heterosis (MPH) for 14 agronomic traits ranged from −5.07% to 106.76%, while the average high-parent heterosis (HPH) ranged from −109.90% to 80.56%. Significant HPH was observed in the growth period (−109.90%), biomass (80.56%), and plant weight (15.20%), which exhibited comprehensive advantages in breeding. Two methods were used to calculate the genetic distance (GD) between the parents based on 2,444,676 single nucleotide polymorphisms (SNPs): total SNPs (GDtotal) and homozygous differential SNPs (GDhomo). Parental clustering based on total SNPs divided the parents into three major types, namely “flat-topped”, “ovate”, and “cylindrical”. These types were more suitable for their traditional pedigree relationships. The correlation between the GD, mean parental phenotype, and heterosis indicated that the phenotype of the parental inbred lines may be useful for predicting heterosis because it is rapid and simple, whereas GD calculated using SNP markers led to a moderate improvement in heterotic prediction. GDhomo was a better predictor of heterosis and F1 performance than GDtotal. These results confirmed that whole-genome SNP markers are an effective tool for evaluating GD in Chinese cabbage and are useful for parent selection in breeding.