Directional epistasis is common in morphological divergence
Cite this dataset
Bourg, Salomé et al. (2024). Directional epistasis is common in morphological divergence [Dataset]. Dryad. https://doi.org/10.5061/dryad.66t1g1k7m
Abstract
Epistasis is often portrayed as unimportant in evolution. While random patterns of epistasis may have limited effects on the response to selection, systematic directional epistasis can have substantial effects on evolutionary dynamics. Directional epistasis occurs when allelic substitutions that change a trait also modify the effects of allelic substitutions at other loci in a systematic direction. In this case, trait evolution may induce correlated changes in allelic effects and effective genetic variance (evolvability) that modify further evolution. Although theory thus suggests a potentially important role for directional epistasis in evolution, we still lack empirical evidence about its prevalence and magnitude. Using a new framework to estimate systematic patterns of epistasis from line-crosses experiments, we quantify its effects on 197 size-related traits from diverging natural populations in 24 animal and 17 plant species. We show that directional epistasis is common and tends to become stronger with increasing morphological divergence. In animals, most traits displayed negative directionality toward larger size, suggesting that epistasis constraints reducing evolvability toward larger size may be common. Dominance was also common and did not systematically alter the effects of epistasis.
README: Directional epistasis is common in morphological divergence
https://doi.org/10.5061/dryad.66t1g1k7m
This folder contains four data files used for the paper "Directional epistasis is common in morphological divergence".
Description of the data and file structure
Dataset – epistasis.csv
This dataset is the raw dataset used in the study "Directional epistasis is common in morphological divergence". The data in this file are not the product of any analysis, they are the results of our literature search. Each row corresponds to one trait. The naming of each derivate follows the diagram "Reciprocal line-cross explanation.pdf". Empty cells indicate the data could not be found.
Study ID : identification name of the study used in our analysis
ID trait : identification number of the trait used in our analysis
Taxa : taxa involved in the line-cross experiment
Cross.type : type of line-cross ; either inter-species or intra-species
Character.category : type of trait; here only morphological trait
Scale : scale of trait ; here only ratio-scale
Measure : type of measure ; here only length (linear trait measurements)
Character : the description of the trait measured
Character_sex : fl (floral trait), no_fl (plant trait not related with the measurement of flowers), sex_related (sex trait in animals), non_sex_related (non sex related trait in animals)
Error.term : type errors reported, here only standard errors (if the error term was reported as variance or standard deviation it has been converted in standard error)
nP1 : number of P1 individuals measured
nP2 : number of P2 individuals measured
nF1_12 : number of F1_12 individuals measured
nF1_21 : number of F1_21 individuals measured
nF2_12 : number of F2_12 individuals measured
nF2_11 : number of F2_11 individuals measured
nF2_22 : number of F2_22 individuals measured
nF2_21 : number of F2_21 individuals measured
nB1_12 : number of B1_12 individuals measured
nB1_11a : number of B1_11a individuals measured
nB1_11b : number of B1_11b individuals measured
nB1_21 : number of B1_21 individuals measured
nB2_21 : number of B2_21 individuals measured
nB2_22 : number of B2_22 individuals measured
nB2_22b : number of B2_22b individuals measured
nB2_12 : number of B2_12 individuals measured
P1 : mean value of the trait measured on P1 individuals
P1error : standard error associated to mean value of P1 individuals
P2 : mean value of the trait measured on P2 individuals
P2error : standard error associated to mean value of P2 individuals
F1_12 : mean value of the trait measured on F1_12 individuals
F1_12error : standard error associated to mean value of F1_12 individuals
F1_21 : mean value of the trait measured on F1_21 individuals
F1_21error : standard error associated to mean value of F1_21 individuals
F2_12 : mean value of the trait measured on F2_12 individuals
F2_12error : standard error associated to mean value of F2_12 individuals
F2_11 : mean value of the trait measured on F2_11 individuals
F2_11error : standard error associated to mean value of F2_11 individuals
F2_22 : mean value of the trait measured on F2_22 individuals
F2_22error : standard error associated to mean value of F2_22 individuals
F2_21 : mean value of the trait measured on F2_21 individuals
F2_21error : standard error associated to mean value of F2_21 individuals
B1_12 : mean value of the trait measured on B1_12 individuals
B1_12error : standard error associated to mean value of B1_12 individuals
B1_11a : mean value of the trait measured on B1_11a individuals
B1_11aerror : standard error associated to mean value of B1_11a individuals
B1_11b : mean value of the trait measured on B1_11b individuals
B1_11berror : standard error associated to mean value of B1_11b individuals
B1_21 : mean value of the trait measured on B1_21 individuals
B1_21error : standard error associated to mean value of B1_21 individuals
B2_21 : mean value of the trait measured on B2_21 individuals
B2_21error : standard error associated to mean value of B2_21 individuals
B2_22 : mean value of the trait measured on B2_22 individuals
B2_22error : standard error associated to mean value of B2_22 individuals
B2_22b : mean value of the trait measured on B2_22b individuals
B2_22berror : standard error associated to mean value of B2_22b individuals
B2_12 : : mean value of the trait measured on B2_12 individuals
B2_12error : standard error associated to mean value of B2_12 individuals
Paperlist – epistasis.csv
This list of studies contains all the papers we have included in our analysis. They were included because they fulfilled our list of requirements described in the supplementary information.
Authors : First authors names of an included study
Year : Year of publication of the study
Study ID : identification name of the study used in our analysis
Journal : Name of the journal where the study is published
Title : Full title of the published study
Sources : from what source this paper came to our knowledge
Derivatives : the set of deritaves that we could use from the study
Natural populations : origins of the parental populations
Comments : relevant comments for the study
Taxa : taxa involved in the line-cross experiment
Species : species involved in the line-cross experiment
Full list of studies – epistasis.csv
This list of studies contains all the papers we have considered through our literature search. As described in Figure S1, 156 were discarded solely on the screening of their title and abstract, while for 886 of them, we assessed the whole manuscript.
Authors : First authors names of a study considered
Year : Year of publication of a study considered
Title : Title of a study considered
Comments : the reason why the paper is or is not included in our study
AICs – epistasis.csv
This file contains the AICs of the five models tested (additive, dominance, additive and dominance, additive and epistasis, additive dominance and epistasis), as presented in the figure S2. Each row corresponds to one trait.
Study ID : identification name of the study used in our analysis
ID trait : identification number of the trait used in our analysis
Add.AIC : AIC from the additive model
Dom.AIC : AIC from the dominance model
Add_dom.AIC : AIC from the additive and dominance model
Pure_epi.AIC : AIC from the additive and epistasis model
Multi.AIC : AIC from the additive, dominance and epistasis model (full model)
Methods
The data were collected through a literature search.
Funding
Norwegian research council (RCN), Award: 287214
Centre for Advanced Study, Award: 223257