Estimates of molecular convergence reveal genes with intermediate pleiotropy underlying adaptive variation across teleost fish
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (Arcadia Science)
Abstract
Molecular convergence, where specific non-synonymous changes in protein-coding genes lead to identical amino acid substitutions across multiple lineages, provides strong evidence of adaptive evolution. Detecting this signal across diverse taxa can reveal broad evolutionary mechanisms that may not be apparent when studying individual lineages. In this study, we search for convergent substitutions in the most speciose group of vertebrates, teleost fishes. Using an unsupervised approach, we detected convergence in 89 protein-coding genes across 143 chromosomal-level genomes. To assess their functional implications, we integrate data on protein properties, gene expression across species and tissues, single-cell RNA sequencing of zebrafish embryonic development, and gene perturbation experiments in zebrafish. The convergent genes were associated with diverse processes including embryonic development, tissue morphogenesis, metabolism, and responses to hormones and heat stress. The convergent substitutions altered amino acid properties, with some occurring at functionally critical sites. Notably, only one-third of these genes were tissue-specific, while the majority were expressed across multiple tissues and cell types. Genetic perturbation data further showed that the convergent genes can affect multiple structures across diverse tissues. These results highlight the pleiotropic nature of the convergent genes. Using an evolutionary modeling approach, we show that adaptive variation tends to accumulate in genes with intermediate pleiotropy, enabling organisms to overcome selective pressures during ecological shifts. Although traditionally considered a source of constraint, we argue that adaptation via genes with intermediate pleiotropy might be particularly advantageous following periods of ecological shifts, and can potentially lead to the evolution of convergent phenotypes.
Article activity feed
-
A one-sided test is a test for directionality. There is no measure of plasticity of perturbation phenotypes available.
-
significantly different from the background (one-sided Kolmogorov-Smirnov test P = 0.011) (Fig 6). Collectively, these findings classify the convergent genes as exhibiting intermediate levels of pleiotropy, supporting previous predictions that moderate pleiotropy facilitates adaptation 63–65.
Is there a significant difference in the mean level of phenotypic plasticity? It's fair to ask whether the distributions of the extent of pleiotropy is the same for convergently vs non-convergently evolved genes, but the conclusions you are reaching about a directional effect (i.e. convergent genes exhibit greater pleiotropy) are not aligned with the tests being conducted, which simply address distributional different, irrespective of direction.
-
-
Thank you for your interest in our preprint.
A well-resolved time-calibrated tree is essential for comparative studies. Constructing phylogenies requires careful attention to multiple aspects of the data such as deciding on the breadth of species or depth of genes. As a result, we used the tree estimated from timetree which synthesises data from multiple publications (where the objective was to infer relationships) to construct the tree.
If we constructed the tree ourselves, any error in the construction would be propagated to the convergence analysis. Although the omegaC metric is robust to topological errors, for the sake of thoroughness we decided to use a phylogeny from timetree as it better representation of our current knowledge of species relationships.
-
We ran the analysis using a rooted time-calibrated species tree obtained from timetree.org 33.
What was the rationale for using a tree from timetree.org as opposed inferring one from the gene families?
I imagine that a comparing the effects of using timetree vs. an inferred tree on CSUBST outputs would be enlightening. Such a comparison could be an empirical way to assess the effects of topological error in this data set (and would be a nice complement to some of the analyses in Fukushima and Pollock).
-