0

I am running an analysis of variances on a large distance matrix using adonis2 as described here: https://www.rdocumentation.org/packages/vegan/versions/2.4-2/topics/adonis

That method is frequently used in microbiome analysis to calculate beta diversity. That's also what I would like to do, i.e. to find out whether my community composition differs in response to an environmental variable (continuous) Permanova returns one p value and there is no "official" post hoc test yet. That's where my question comes in:

I've come across publications saying they adjusted their permanova result using FDR/BH method. I cannot wrap my head around this. I'm confident I understand how FDR correction is calculated, I just don't see how that would be done for PERMANOVA, or, even more, how I would code it.

Can anyone help me out here?

mucl
  • 113
  • 1
  • 6

1 Answers1

0

Would be clearer if you provide an example of so-called publication. You are right that for each variable, permanova returns 1 p-value. However, if the model includes many variables, you would have 1 p-value for each variable and you need to correct for FDR.

For example in this publication looking at variation in gut microbiome, they wrote:

To calculate the variation explained by each of our collected host factors, we performed an Adonis test implemented in QIIME. Each host factor was calculated according to its explanation rate, and P values were generated based on 1,000 permutations. All P values were then adjusted using the Benjamini–Hochberg method.

You can also see an example of this in Table S2, I attached a screenshot here:

enter image description here

StupidWolf
  • 45,075
  • 17
  • 40
  • 72
  • https://doi.org/10.1016/j.chom.2019.07.004 and https://jrnlclub.org/research-films/obesity-t2dm-microbiome (minute 5:11) I too thought it was related to confounders, but they somehow obtain a q value when not correcting for anything. – mucl Aug 18 '20 at 11:57
  • also unrelated but did I misuse the term "publication" or what are you referring to? – mucl Aug 18 '20 at 12:01
  • not clear what they did. i can only suppose each column is a different phenotype and each row represents the different terms in the model. sorry not sure what is your question anymore. My point is, if you fit a model with more than 1 term, you do the FDR correction with the p-values, like i showed in the example – StupidWolf Aug 18 '20 at 13:43
  • If something else was done, it would be unconventional and you would have to ask the authors. – StupidWolf Aug 18 '20 at 13:44
  • Thanks for your comments. Just to clarify one more presumably basic question: say I calculate adonis10 times, each model is community~variable (just one variable, like eg BMi, I get a list of 10 p values, would I have to correct for multiple testing? – mucl Aug 18 '20 at 17:30
  • If you are always testing the dependent variable (community), answer is yes. – StupidWolf Aug 18 '20 at 19:38