Elmer & Meyer 2011: Adaptation in the age of ecological genomics: insights from parallelism and convergence

Natural selection is one of the two major forces which drive evolution of species, morphs and phenotypes. However, due to the confounding effects of environmental stochasticity, replication at the taxon level is needed for better understanding the influence and importance of natural selection in evolutionary biology. Parallel evolution events, in which related taxons independently evolve similar traits, provide a useful framework to investigate the mechanisms of adaptation using powerful new genomic and transcriptomic tools.

In the paper “Adaptation in the age of ecological genomics: insights from parallelism and convergence”, Kathryn Elmer and Axel Meyer reviewed examples of parallel evolution in natural populations of non-model species and compared the genetic bases of their adaptive traits. Inspired by the hypotheses that parallel phenotypes share homologous genetic bases, they investigated the advances allowed by new genomic technique in the field of adaptive evolution. Understanding genetic origins and mechanisms of phenotypic changes will raise insights into the opportunities for species to adapt under ecological pressure.

The authors proposed a classification for the nature of genetic variations leading to similar phenotypes among three levels: homologous mutation at the same nucleotides, homologous mutation in the same gene at different nucleotide and non-homologous mutation in different genes. Mutations participating to any of these categories can be part of the genetic standing variation, the pool of old mutations already present in ancestral population or may have appeared de novo after parallel evolution begun.

The new emerging genomic methods will be very useful in identifying variation responsible for adaptation because they allow broad analyses at the population level without a priori hypotheses, unlike the older but reliable methods focusing on candidate genes. Efficient mapping of phenotypes now permits identification of genome parts involved, and loci under selection can be tracked through genome scans.

The compilation of studies using molecular and geographical wide methods on different species or complexes revealed that parallel evolution of phenotypes is driven by all categories of mutations, at same or different genes. A representative example is provided by studies of coat coloration in mice species of genus Peromyscus. Unless these first results need to be supported by other studies, it suggests that a broad variety of genetic mechanisms may are responsible for parallel evolution and a clear pattern is still to emerge. Despite this large evolutionary potential, the phenotypic response seems limited by morphological and developmental constraints, suggesting there is no tight couple between genetic bases and phenotypes. Until now, the focus on this king of mapping between genotypes and phenotypes may have clouded the genetic variability newly emphasised by genomics methods.

Accordingly with authors’ view, our discussion firstly focused on the absence of clear pattern revealed by the review. New genomic methods are emerging and have been only applied a few times in studies of parallel adaptation. In fact, nearly all studies carried so far are reviewed here, laying the basis for future research. The conclusions, emphasizing the complexity of mechanisms, raised questions about the pertinence of knowing precisely which genes are responsible for adaptation. However, the main question the authors set does not much concern the proximal mechanisms, but the complexity of these mechanisms: a constant pattern in the causes driving adaptation may be interpreted as a determined strategy of Nature allowing for the evolution of parallel phenotypes. On the other hand, random mechanisms would reflect an important role of stochasticity in this process of evolution.

The proposed classification for nucleotides changes discriminates between mutations happening in same or different genes. This last option potentially affects phenotypes through an extensive amount of mechanistic patterns of expression and/or regulation, in contrast to the other categories. We thus discussed the relevance of splitting it in two according that the mutated genes have similar functions or not. Also debated was the relevance of grouping the two categories of mutations happening on the same gene, considered as unlikely to happen in parallel evolution. However, homologous mutation at the same nucleotide would be more likely to be due to standing genetic variation than to de novo mutation, and it is worth making the difference.

The possibilities of building tests for the basis of parallel adaption were discussed, in light of the experiment about foxes’ domestication conducted in Siberia during the last 50 years. Such long lasting experiment reveals that evolution may be relatively quick under stringent selective conditions in evolved animals. However, more realistic tests would preferentially use bacteria to produce results quicker.

I am personally not familiar with new genomics methods, having a more traditional background of searching for candidate genes. The main impact of the review is thus emphasizing on the power of these democratizing methods. I believe it is important to focus researchers’ attention on emerging new methodologies as soon as possible in order to boost advances in comprehension of evolution.

Elmer, K., & Meyer, A. (2011). Adaptation in the age of ecological genomics: insights from parallelism and convergence Trends in Ecology & Evolution, 26 (6), 298-306 DOI: 10.1016/j.tree.2011.02.008