Population structure and signals of local adaptation in Eugenia uniflora (Myrtaceae), a widely distributed species in the Atlantic Forest

Evolutionary diversity in species can arise in many ways, including local adaptation. Despite the global importance of tropical forest ecosystems, few studies have explored patterns of local adaptation in tropical tree species. We investigated population genetic structure and adaptive genetic diversity in Eugenia uniflora, a widely distributed tree species across the Atlantic Forest Domain (AFD). We sampled E. uniflora in distinct environments from the AFD and genotyped 523 single-nucleotide polymorphisms (SNPs) using a candidate gene approach. We studied neutral population structure and used two outlier methods based on population differentiation (Bayescan and LEA) and one gene-environment association analysis (LFMM) to detect putative SNPs under divergent selection. We detected spatial population structuring between the northern and southern range of the species distribution and higher genetic diversity in southern populations from the riparian forest. We found 18 candidate loci potentially involved in local adaptation, among which LEA and LFMM both detected seven loci. Overall, precipitation-related variables were more represented in statistically significant genotype-climate associations (c. 60%). Our study provides a first insight into the distribution of adaptive genetic variation in E. uniflora, highlighting how tropical tree species may adapt over time and across the AFD.