Abstract
Amino acid propensities at a site change in the course of protein evolution. This may happen for two reasons. Changes may be triggered by substitutions at epistatically interacting sites elsewhere in the genome. Alternatively, they may arise due to environmental changes that are external to the genome. Here, we design a framework for distinguishing between these alternatives. Using analytical modelling and simulations, we show that they cause opposite dynamics of the fitness of the allele currently occupying the site: it tends to increase with the time since its origin due to epistasis (“entrenchment”), but to decrease due to random environmental fluctuations (“senescence”). By analysing the genomes of vertebrates and insects, we show that the amino acids originating at negatively selected sites experience strong entrenchment. By contrast, the amino acids originating at positively selected sites experience senescence. We propose that senescence of the current allele is a cause of adaptive evolution.
Highlights
The magnitude of k was overestimated for simulations with accelerated evolution rate, the estimates were not biased in any direction (t-test p value = 0.53, Supplementary Figs. 6 and 7, top panel; Supplementary Table 2)
We found that ancestral states reconstruction slightly biased both k and the fraction of alleles with changing fitness upwards, but the confusion frequency remained low (0% for the two-parameter model, 0.5% for the three-parameter model), and the only erroneously classified simulation had a low fraction of entrenched alleles (0.09)
While the shortest trees don’t provide the time range sufficient to detect SPFL changes, the largest ones lack resolution and result in bigger prediction error
Summary
From these alignments, we reconstructed the alleles in the internal nodes of phylogenetic trees with codeml[53]. We re-estimated the lengths of individual branches as the average frequency of amino acid substitutions per site on this branch. We classified codon sites as negatively selected (ω < 1), neutral (ω = 1), or positively selected (ω > 1) using the Bayes empirical Bayes method as implemented in the PAML package[54]. The mitochondrial dataset consists of the amino acid alignment of five proteins for several thousand metazoan species[14]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
