Cargando…
The Role of the Effective Population Size in Compensatory Evolution
The impact of the effective population size (N(e)) on the efficacy of selection has been the focus of many theoretical and empirical studies over the recent years. Yet, the effect of N(e) on evolution under epistatic fitness interactions is not well understood. In this study, we compare selective co...
Autores principales: | , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Oxford University Press
2011
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3140890/ https://www.ncbi.nlm.nih.gov/pubmed/21680889 http://dx.doi.org/10.1093/gbe/evr057 |
Sumario: | The impact of the effective population size (N(e)) on the efficacy of selection has been the focus of many theoretical and empirical studies over the recent years. Yet, the effect of N(e) on evolution under epistatic fitness interactions is not well understood. In this study, we compare selective constraints at independently evolving (unpaired) and coevolving (paired) sites in orthologous transfer RNAs (tRNA molecules for vertebrate and drosophilid species pairs of different N(e). We show that patterns of nucleotide variation for the two classes of sites are explained well by Kimura's one- and two-locus models of sequence evolution under mutational pressure. We find that constraints in orthologous tRNAs increase with increasing N(e) of the investigated species pair. Thereby, the effect of N(e) on the efficacy of selection is stronger at unpaired sites than at paired sites. Furthermore, we identify a “core” set of tRNAs with high structural similarity to tRNAs from all major kingdoms of life and a “peripheral” set with lower similarity. We observe that tRNAs in the former set are subject to higher constraints and less prone to the effect of N(e), whereas constraints in tRNAs of the latter set show a large influence of N(e). Finally, we are able to demonstrate that constraints are relaxed in X-linked drosophilid tRNAs compared with autosomal tRNAs and suggest that N(e) is responsible for this difference. The observed effects of N(e) are consistent with the hypothesis that evolution of most tRNAs is governed by slightly to moderately deleterious mutations (i.e., |N(e)s| ≤ 5). |
---|