I'm attempting to run RDA on a SNP dataset to search for effects of isolation-by-distance (IBD). The script I have now (Meirmans, 2015: Article Seven common mistakes in population genetics and how to avoid them
) takes geographic coordinates and transforms them. After some pushback from reviewers, I decided to calculate pairwise geographic distances between sampling localities using marmap (R package).Is there a way to run RDA on a SNP dataset with a matrix of pairwise geographic distances? I can find online guides and practice sets that deal with environmental variables, but I'm having trouble figuring out how to apply these examples to pairwise geographic distances. Is this even appropriate?
I'm running Mantel tests as well. That's been easy-peasy.
I think it depends on how you configure your datasets and what you want to answer, you can construct a dataset with your "variables" being the pairwise geographic distance but you may need more variables to run the RDA. Why are you running an RDA in the first place instead of a PCA or a heatmap? RDA is useful in detecting putatively outliers SNPs related to variables/environmental conditions. You can consult this paper, this may be helpful. https://www.nature.com/articles/s41437-021-00494-x
- Badges
- Science topic