Bird genetic diversity
Motivation
Ecologists and evolutionary biologists have long investigated links between diversity and geography. For example, as the area of a habitat increases, the number of different species found within it increases—a seemingly obvious finding, but the exact shape of the relationship has been found to relate to the rates of immigration and extinction in the habitat, predator-prey behaviors, and so on.
A similar relationship would seem likely to exist within a species. A species that lives over a small area might have low genetic diversity, because individuals frequently interbreed and there is little opportunity for some individuals to evolve separately from others. A species that lives over a wide area might have high genetic diversity, both because the varied habitat in which it lives means different individuals experience different natural selection pressures, and because individuals far apart may never be able to interbreed.
The exact nature of this relationship could be used to test hypotheses about evolution and natural selection. This dataset comes from a paper that explored the relationship in birds, which have widely varying geographic ranges and hence could be a good group to work with.
Data
This data includes 387 bird species, along with two measures of their genetic diversity, a measure of the size of their breeding range, their body mass, and whether they are migratory or resident birds. The breeding ranges vary in size from a just few islands in the Seychelles, for the Seychelles warbler (Acrocephalus sechellensis), to enormous ranges of the Southern Ocean for the grey-headed albatross (Thalassarche chrysostoma).
Data preview
bird-diversity.csv
Variable descriptions
| Variable | Description |
|---|---|
| Species | Bird species name |
| Family | Family of birds to which this species belongs |
| Heterozygosity | A measure of genetic richness. Birds (and most animals) have two copies of each chromosome; expected heterozygosity is the probability that for a randomly selected gene, these two copies will be different versions (alleles) of the gene. |
| Allelic richness | The average number of different alleles (versions of a gene) for each gene. |
| Breeding range size | Size of the spatial range over which this bird breeds, in units of 10,000 square km |
| Body mass | Average body mass of the species (grams) |
| Latitude | Mid-point latitude of the species breeding range. (All latitudes are positive, even those in the southern hemisphere.) |
| Migratory status | Is this species migratory or does it stay resident in one area? |
| References | References for the data for this species |
Questions
- Body size and latitude are thought to be related to genetic diversity for reasons unrelated to breeding range size. Regress heterozygosity against breeding range size, body mass, and latitude. Examine the shape of the relationship, which is thought not to be linear and suspected to have an asymptote. Does the relationship depend on whether the species are migratory? Compare this to the shape you find using allelic richness instead of heterozygosity as the outcome.
- Species in this dataset are grouped into families; by definitions, species within a family are more similar than species in two different families. This may have effects on the assumptions of the regression. What assumptions may be violated, and what effect would this have on your results? (Note that the original paper accounted for this effect by taking into account the species’ phylogenetic trees.)
- Fit a hierarchical model to account for this effect, using family to define the hierarchy. How do your results change?
References
Fan H, Zhang Q, Rao J, Cao J, Lu X (2019) Genetic diversity-area relationships across bird species. The American Naturalist, volume 194, number 5. https://doi.org/10.1086/705346
Fan H, Zhang Q, Rao J, Cao J, Lu X (2019) Data from: Genetic diversity-area relationships across bird species. Dryad Digital Repository. https://doi.org/10.5061/dryad.9p380m9