Genetic & Phylogenetic structure

Commands in mothur
1

Hi,

I have been applying the uncorrected p-distance, as implemented in pairwise.dist() followed by the workflow giving me the Jaccard dissimilarity index for different cut-offs, as a proxy for genetic differentiation among communities (then analysed by the AMOVA). In addition to this, I have used the unweigthed UniFrac to represent the phylogenetic structure of these communities.

However, Im not totally convinced that these analyses are really telling me the things Im claiming to have analysed. Is there any point of doing this, in the sense Im doing it, since everything is based on the same alignment? If Im analysing the alignment (with some distance matric) could I say this would be on a more fine-scaled level than if I would analyse a tree (with some distance metric)?

Sorry if posting question in wrong section.

Thanks
Johannes

2

I have been applying the uncorrected p-distance, as implemented in pairwise.dist() followed by the workflow giving me the Jaccard dissimilarity index for different cut-offs, as a proxy for genetic differentiation among communities (then analysed by the AMOVA). In addition to this, I have used the unweigthed UniFrac to represent the phylogenetic structure of these communities.

You are putting your distances through an OTU filter so AMOVA isn’t really comparing genetic differentiation in this case - it’s comparing the similarity of community memberships. What you perhaps want to do is to run the output of pairwise.dist through amova. Unfortunately, I suspect this may be computationally onerous…
Pat