I have 6 samples. 3 for treatment and 3 for control. My PCoA plot has clear and very large differential clustering (https://imgur.com/k6ilAU0). But ADONIS test gives a p-value of 0.1. My understanding is that I have got too few sample for the adonis test. What test should I use to check for dissimilarity in such small pool of samples and what are the command lines for that in QIME2? Thank you
1 Like
Hi @ehossain,
You are correct in that you have too few samples, this fact is going to be true no matter what test you use. For example, you can also use the q2-diveristy beta group significance action but I doubt you would see anything different. I’ll go one step further and say that even if you did see a p value <0.05 this is unfortunately entirely unreliable and essentially meaningless with n=3. I personally think you should just not bother with a statistical test here and let that image speak for itself with an explanation.
1 Like
Ya, I agree. Thank you for the verification…that is exactly what I am gonna tell.
1 Like