Lack of signals of selection at candidate loci at a small geographical scale along a steep altitudinal gradient in Norway spruce (Picea abies [L.] Karst.)

Abstract

Local adaptation is a key concept in biology: shift of genetic structures of populations due to differential survival of genotypes is expected to lead to phenotypes providing an advantage in the local environment. Variation of sequences of twelve candidate genes was investigated in 13 Norway spruce (Picea abies (L.) Karst.) provenances originating from sites distributed along an altitudinal gradient from 550 to 1300 m a.s.l. Signals of selection were assessed in 103 single nucleotide polymorphisms (SNP). The Bayesian FST-outlier identification methods as implemented in the programs BayeScan and Arlequin did not identify any SNP with a clear evidence of selection. The approaches relying on SNP-climate associations (spatial analysis method based on logistic regression of allele frequencies with environmental variables, Bayesian method applied in BayEnv2) identified several relationships but none of them remained significant after correction for multiple testing. Gene flow, epigenetic inheritance and former management of the studied populations are discussed as potential reasons for this weak evidence of selection signals.