Landscape genomics on Acrocomia aculeata trees aiming breeding selection

Abstract Acrocomia aculeata is a promising oil-producing palm whose breeding is constrained by its long juvenile period, high costs of phenotypic evaluations, and limited genomic resources. Here, we present an integrative framework combining landscape genomics and genomic selection to support breeding strategies in A . aculeata . Spatial and genomic information from 167 individuals, distributed across four provenances, was analyzed using three SNP-calling strategies: a de novo pipeline, the Elaeis guineensis reference genome, the A. aculeata transcriptome; and a combined SNP dataset. Population structure and genetic diversity were evaluated across provenances. Genomic selection models were developed for pulp dry mass (PDM), kernel dry mass (KDM), and oil content (OC) using leave-one-out and provenance-based validation. Population analyses revealed low genetic differentiation among provenances (F ST = 0.001–0.003), indicating that the trees constitute a single population. Genomic prediction achieved predictive abilities of up to 0.35 for PDM, 0.43 for KDM, and 0.50 for OC with the oil palm genome and combined SNP datasets showing the most consistent predictive performance. Incorporating a spatial kernel showed that environmental effects accounted for up to 69.32% of the phenotypic variance in OC. Spatial genomic prediction identified superior genotypes for simultaneous improvement of the traits , which were distributed throughout the landscape rather than concentrated within specific provenances. Overall, our results suggest that seed collection based on individually superior trees may be more effective than collection based solely on geographic provenance and highlight the potential of integrating genomic prediction to exploit natural genetic diversity and accelerating the breeding of native species.