Population genomic structure and genome-wide linkage disequilibrium in farmed Atlantic salmon (Salmo salar L.) using dense SNP genotypes

Abstract

AbstractChilean Farmed Atlantic salmon (Salmo salar) populations were established with individuals of both European and North American origins. These populations are expected to be highly genetically differentiated due to evolutionary history and poor gene flow between ancestral populations from different continents. The extent and decay of linkage disequilibrium (LD) among single nucleotide polymorphism (SNP) impacts the implementation of genome-wide association studies and genomic selection and provides relevant information about demographic processes of fish populations. We assessed the population structure and characterized the extent and decay of LD in three Chilean commercial populations of Atlantic salmon with North American (NAM), Scottish (SCO) and Norwegian (NOR) origin. A total of 151 animals were genotyped using a 159K SNP Axiom® myDesign™ Genotyping Array. A total of 40K, 113K and 136 K SNP markers were used for NAM, SCO and NOR populations, respectively. The principal component analysis explained 86.7% of the genetic diversity between populations, clearly discriminating between populations of North American and European origin, and also between European populations. Admixture analysis showed that the Scottish and North American populations likely come from one ancestral population, while the Norwegian population probably originated from more than one. NAM had the lowest effective size, followed by SCO and NOR. Large differences in the LD decay were observed between populations of North American and European origin. A r2 threshold of 0.2 was estimated for marker pairs separated by 8,000 Kb, 42 and 64 Kb in the NAM, NOR and SCO populations, respectively. In this study we show that this SNP panel can be used to detect association between markers and traits of interests and also to capture high-resolution information for genome-enabled predictions. Also, we suggest the feasibility to achieve higher prediction accuracies by using a small SNP data set as was used with the NAM population.