Genetic and metabolomic differentiation of Dryas octopetala in Svalbard: genome-wide SNP evidence and region-specific marker development

Background Dryas octopetala , the dominant tundra species in Svalbard, is a suitable model for examining population differentiation under extreme environmental conditions. Although earlier amplified fragment length polymorphism and simple sequence repeat studies have provided baseline information, the genomic and metabolomic structures within the Svalbard populations have not been comprehensively characterized. Therefore, we aimed to investigate the genetic structure of D. octopetala populations across seven sites in Svalbard using genome-wide single nucleotide polymorphism (SNP) genotyping. Results Genotyping-by-sequencing of 33 individuals yielded 33,010 high-quality SNPs, revealing moderate but significant population differentiation across seven sites (Φ ST = 0.154, p = 0.0001), with 84.6% of total genetic variation occurring within populations. ADMIXTURE identified two major regional clusters, separating Longyearbyen and Ny-Ålesund. Principal component analysis and discriminant analysis of principal components similarly distinguished individuals from these two regions, confirming a consistent regional-level genetic structure. Subsequently, 13 polymerase chain reaction (PCR)-based SNP markers were developed from 150 polymorphic SNPs that differentiated these regions, which showed stable amplification and reliably reproduced regional genetic patterns with high allelic richness and heterozygosity in the Ny-Ålesund population. Metabolomic profiling using gas chromatography–mass spectrometry and liquid chromatography–mass spectrometry detected 1,466 peaks, of which 33 metabolites differed significantly between the regions. Orthogonal partial least squares-discriminant analysis clearly separated the Longyearbyen and Ny-Ålesund samples, with compounds such as adenosine- and galactose-related metabolites contributing to this distinction. Mantel and Procrustes analyses revealed a significant correlation between genomic and metabolomic distance matrices. Conclusions This study provides integrated genomic and metabolomic evidence of regional differentiation in D. octopetala across Svalbard and demonstrates that the newly developed PCR-based SNP markers can effectively distinguish regional populations while reflecting genome-wide patterns of population differentiation. The concordance between genetic and metabolomic patterns highlights the value of multi-omics approaches for characterizing regional population differentiation and associated functional variation in Arctic plants.