Combining climatic and genomic data improves range-wide tree height growth prediction in a forest tree

Population response functions based on climatic and phenotypic data from
common gardens have long been the gold standard for predicting
quantitative trait variation in new environments. However, prediction
accuracy might be enhanced by incorporating genomic information that
captures the neutral and adaptive processes behind intra-population
genetic variation. We used five clonal common gardens containing 34
provenances (523 genotypes) of maritime pine (Pinus pinaster Aiton) to
determine whether models combining climatic and genomic data capture the
underlying drivers of height-growth variation, and thus improve
predictions at large geographical scales. The plastic component explained
most of the height-growth variation, probably resulting from population
responses to multiple environmental factors. The genetic component stemmed
mainly from climate adaptation, and the distinct demographic and selective
histories of the different maritime pine gene pools. Models combining
climate-of-origin and gene pool of the provenances, and positive-effect
height-associated alleles (PEAs) captured most of the genetic component of
height-growth and better predicted new provenances compared to the
climate-based population response functions. Regionally-selected PEAs were
better predictors than globally-selected PEAs, showing high predictive
ability in some environments, even when included alone in the models.
These results are therefore promising for the future use of genome-based
prediction of quantitative traits.