Data from: probing variation in reaction norms in wild populations: the importance of reliable environmental proxies
Many traits are phenotypically plastic, i.e., the same genotype expresses different phenotypes depending on the environment. Genotypes and individuals can vary in their response to the environment and this genetic (G×E) and individual (I×E) variation in reaction-norm slopes can have important ecolog...
| Main Authors: | , , , |
|---|---|
| Format: | Dataset |
| Language: | unknown |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://zenodo.org/record/8360141 https://doi.org/10.5061/dryad.zcrjdfnjr |
| Summary: | Many traits are phenotypically plastic, i.e., the same genotype expresses different phenotypes depending on the environment. Genotypes and individuals can vary in their response to the environment and this genetic (G×E) and individual (I×E) variation in reaction-norm slopes can have important ecological or evolutionary consequences. Studies on I×E/G×E often fail to show slope variation, potentially due to the choice of the environmental covariate. Identifying the genuine environmental driver of phenotypic plasticity (the cue) is practically impossible and hence only proxies can be used. If the proxy is too weakly correlated with the cue, this may lead researchers to conclude there is little or no (variation in) plasticity, and hence lead to downwardly biased estimates of the potential for plastic responses (or evolutionary change in the slope) in response to environmental change. Alternatively, the Environment-Specific Mean phenotype (ESM) across individuals—which captures all environmental effects on the phenotype—as covariate should be less prone to such bias. We showed by simulation—after verifying the concept analytically—that using weakly correlated proxies indeed biased estimates of slope variation vis-à-vis the true cue downward but that ESM as a covariate held up well, even when multiple sources of I×E or an interaction between environments (I×E×E) existed in the data. Analysis of two real datasets revealed that estimated I×E and G×E, respectively, were more sizeable and precise when using ESM as opposed to reasonably informative environmental proxies. We argue that the ESM approach should be adopted by biologists as a yardstick in the study of (variation in) plasticity in the wild and that it may serve as a useful starting point for the search of better environmental proxies and unravelling complex I×E or G×E patterns. Data contain phenotypic breeding data and NAO data for the Common guillemot as described in Reed et al. (2006). These data were used as one of the two practical examples in the paper. ... |
|---|