Estimating evolutionary potential in the wild : Role and stability of the G matrix

Background/Question/Methods

Global environmental change represents a major threat to biodiversity today. The habitats of species are being modified at a fast rate, raising an important question: will populations manage to adapt to the new conditions? One strong constraint on microevolutionary responses is the fact that traits within an organism are not free to evolve independently, but form integrated units. From an adaptive point of view, this ensures that traits can function together, but can prevent response to selection.

Here we aim to quantify constraints on evolution. This requires assessing (1) how frequently evolutionary responses are constrained by genetic correlations, i.e.to which extent the rate of adaptation is decreased by genetic correlations and (2) how stable these constraints are, i.e. if genetic correlations are changing rapidly, the G matrix (the matrix of genetic variance-covariances, summarizing the evolutionary potential) should not represent a strong constraint on evolutionary responses

These questions can be answered long term data sets of wild bird populations for which detailed pedigrees are available. These data allow to assess the G matrix and selection gradient with associated confidence interval so that these errors can be propagated to the next step of analyses (prediction of response to selection, comparisons among matrices).

Results/Conclusions

Using a comparative approach across seven bird species, we found that the rate of adaptation of morphological traits was decreased on average by 25% because of genetic correlations. This result is of peculiar interest because morphological traits are often thought to be submitted to low evolutionary constraints because of their high heritabilities. Our results emphasize the importance of taking into account multivariate constraints when investigating rapid adaptation in changing environments. These constraints arise because while all genetic correlations are positive, selection can favour an increased value (positive directional selection) for some trait and a decreased value for some other trait (negative directional selection) i.e. some unlikely combinations of traits are favored, making adaptation difficult. This could have important impacts on evolutionary dynamics and hence population persistence in the face of rapid environmental change if the constraints are stable.

The stability of the G matrix was investigated in terms of orientation and overall evolvability. The stability of these parameters depends on the temporal scale (generation time vs. longer than generation time). Overall results suggest various levels of evolutionary constraints among species.