Development of the rubber tree 40K breeding chip with applications in genetic study and breeding prediction

As an important industrial crop, rubber tree ( Hevea brasiliensis ) provides the source of natural rubber globally. However, rubber tree breeding is confronted with the challenge of long breeding cycles and corresponding time, land, and labor costs. In this study, we developed a novel high-density breeding chip for rubber tree using genotyping by target sequencing (GBTS) technology, comprising approximately 40 K markers. The developed breeding chip was used to genotype 223 rubber tree clones. Results showed a prevalent low genetic diversity among the rubber tree clones, indicating a limited genetic foundation. Subsequently, markers associated with girth growth rate were identified through genome-wide association study (GWAS). Based on the GWAS results, seven genomic selection (GS) models were used to assess the prediction accuracy of different associated marker datasets for girth growth rate. The Bayesian least absolute shrinkage and selection operator (BayesLASSO), using the marker dataset with a p -value < 0.001, showed the highest prediction accuracy of 0.59, thus being selected as the optimal GS model and genotype dataset for breeding prediction. Fifty superior parental combinations were selected by usefulness criteria calculated based on genome estimated breeding values of simulated F 1 progenies. This study developed a reliable breeding chip applicable in genetic study and breeding prediction and provided a strategy for molecular breeding of rubber trees by determining suitable parental combinations without the need for extensive field trials. • Development of the first high-density rubber tree 40K breeding chip. • Diversity analysis revealed a prevalence of low genetic diversity in rubber tree clones. • Enhancing genomic selection prediction accuracy by combining GWAS results. • Screened superior potential parental combinations through breeding prediction.