Conifers are characterized by a large genome size and a rapid decay of linkage disequilibrium, most often within gene limits. Genome scans based on noncoding markers are less likely to detect molecular adaptation linked to genes in these species. In this study, we assessed the effectiveness of a genome-wide single nucleotide polymorphism (SNP) scan focused on expressed genes in detecting local adaptation in a conifer species. Samples were collected from six natural populations of white spruce (Picea glauca) moderately differentiated for several quantitative characters. A total of 534 SNPs representing 345 expressed genes were analysed. Genes potentially under natural selection were identified by estimating the differentiation in SNP frequencies among populations (F ST ) and identifying outliers, and by estimating local differentiation using a Bayesian approach. Both average expected heterozygosity and population differentiation estimates (H E = 0.270 and F ST = 0.006) were comparable to those obtained with other genetic markers. Of all genes, 5.5% were identified as outliers with F ST at the 95% confidence level, while 14% were identified as candidates for local adaptation with the Bayesian method. There was some overlap between the two gene sets. More than half of the candidate genes for local adaptation were specific to the warmest population, about 20% to the most arid population, and 15% to the coldest and most humid higher altitude population. These adaptive trends were consistent with the genes' putative functions and the divergence in quantitative traits noted among the populations. The results suggest that an approach separating the locus and population effects is useful to identify genes potentially under selection. These candidates are worth exploring in more details at the physiological and ecological levels.

Spatial climate models were developed for México and its periphery (southern USA, Cuba, Belize and Guatemala) for monthly normals of average, maximum and minimum temperature and precipitation using thin plate smoothing splines of ANUSPLIN software on ca. 3,800 observations. The fit of the model was generally good: the signal was considerably less than one-half of the number of observations, and reasonable standard errors for the surfaces would be less than 1 • C for temperature and 10-15% for precipitation. Monthly normals were updated for three time periods according to three General Circulation Models Climatic Change (2010) 102:595-623 and three emission scenarios. On average, mean annual temperature would increase 1.5 • C by year 2030, 2.3 • C by year 2060 and 3.7 • C by year 2090; annual precipitation would decrease −6.7% by year 2030, −9.0% by year 2060 and −18.2% by year 2090. By converting monthly means into a series of variables relevant to biology (e. g., degree-days > 5 • C, aridity index), the models are directly suited for inferring plant-climate relationships and, therefore, in assessing impact of and developing programs for accommodating global warming. Programs are outlined for (a) assisting migration of four commercially important species of pine distributed in altitudinal sequence in Michoacán State (b) developing conservation programs in the floristically diverse Tehuacán Valley, and (c) perpetuating Pinus chiapensis, a threatened endemic. Climate surfaces, point or gridded climatic estimates and maps are available at http://forest.moscowfsl.wsu.edu/climate/.

Genomic selection (GS) is of interest in breeding because of its potential for predicting the genetic value of individuals and increasing genetic gains per unit of time. To date, very few studies have reported empirical results of GS potential in the context of large population sizes and long breeding cycles such as for boreal trees. In this study, we assessed the effectiveness of marker-aided selection in an undomesticated white spruce (Picea glauca (Moench) Voss) population of large effective size using a GS approach. A discovery population of 1694 trees representative of 214 open-pollinated families from 43 natural populations was phenotyped for 12 wood and growth traits and genotyped for 6385 single-nucleotide polymorphisms (SNPs) mined in 2660 gene sequences. GS models were built to predict estimated breeding values using all the available SNPs or SNP subsets of the largest absolute effects, and they were validated using various cross-validation schemes. The accuracy of genomic estimated breeding values (GEBVs) varied from 0.327 to 0.435 when the training and the validation data sets shared half-sibs that were on average 90% of the accuracies achieved through traditionally estimated breeding values. The trend was also the same for validation across sites. As expected, the accuracy of GEBVs obtained after cross-validation with individuals of unknown relatedness was lower with about half of the accuracy achieved when half-sibs were present. We showed that with the marker densities used in the current study, predictions with low to moderate accuracy could be obtained within a large undomesticated population of related individuals, potentially resulting in larger gains per unit of time with GS than with the traditional approach.