I have a metagenomic dataset originating from the gut contents of a marine mollusk. I am trying to identify the appropriate statistical workflow for analysing these data. While I do have abundance data, it is based on read count, which is a notoriously inaccurate measure of abundance due to numerous factors. Due to this, we are opting to use presence/absence data to focus more on the frequency of occurrence of the different taxa across our metadata.
I've produced a Jaccard distance matrix, specifying binary data, using `vegan` in R. Using that matrix I've performed a permutational multivariate analysis of variance (adonis), though its slightly dubious, as I am receiving significant outputs on some metadata that don't appear to have any visual differences on NMDS plots produced with the same Jaccard matrix. Now I fear I might be completely barking up the wrong tree, as stats are not my strong point.
Would anyone with experience analysing these kinds of data be able to point me in the right direction so I know I'm conducting the correct analysis?
Thank you very much!
- Badges
- Science method
- Similar topics
- Biological Science
- Ecology