I am new to stats to this level in ecology. I am trying to compare DNA and RNA libraries with thousands of OTUs. I summarized taxa to get the most abundant species, but I can obtain only relative abundances. I was thinking to use SIMPER as I read in several comments to test which species differ the most per station between DNA and RNA based libraries. However I read that SIMPER is a more or less robust test. I was wondering if the manyglm was also an alternative for my question or if you suggest another way. Thank you for your help!
It may help to see the structure of your data. SIMPER can be difficult, mostly because it can be difficult to interpret (especially if you have many contrasts). However, I suggest giving it a shot. It won't give you a p-value, but it will give you an idea what species drive the differences between your stations. It may show you relationships that you can follow up with a more hypothesis-driven test.
That said, PERMANOVA (adonis in the vegan package) is a popular test that will help you determine if the dissimilarities between stations are significant. If you find significant dissimilarities, the SIMPER will help you identify which species are the likely culprits.
Good luck!
An indicator species analysis (IndVal) will inform you about the characterisitc OTU's if you compare groups (labdsv package):
Dufrêne, M. and P. Legendre, 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecological Monographs, 67 : 345-366.
Thanks for both your answers, sorry i was away for research purpose. I changed orientations and I selected major taxa of my libraries (thus 18 different taxa) with their relative abundances (still for both DNA and RNA). I will apply your method Daniel and give you feedback! Thanks also Klaus for your suggestion. but I am not sure it is corresponding to what I look for. I am trying to compare one sample on its DNA and RNA libraries, and then identify the taxa that are the most influencing the differences (like SIMPER does). Or I may have missunderstood the article?
As Rion Elias Lerm points out above mvabund/manyglm will likely be a better approach than SIMPER
SIMPER has the downside of not being able to identify species which are driving the differences unless there is a high degree of variance in the species in question. See "Distance-based multivariate analyses confound location and dispersion effects" by Warton et al. for a more detailed explanation.
- Badges
- Science method
- Similar topics
- Biological Science
- Ecology