Hello qiime2 universe!
I’m using the q2 longitudinal plugin for the first time, as I was looking for a solution for repeated measures analysis of beta diversity data (over 3 timepoints) and understood that PERMANOVA was not going to cut it.
The pipeline I am doing is this:
-
qiime longitudinal first-distances -
qiime longitudinal linear-mixed-effects
The piece that I am uncertain about is that I want the individual ID to be the random effect, but I am uncertain if I add that under --p-random-effects or not.
In the help it says the following:
--p-random-effects TEXT Comma-separated list (without spaces) of metadata
columns to use as independent covariates used to
determine the variance and covariance structure
(random effects) of "metric". To add a random slope,
the same value passed to "state-column" should be
passed here. A random intercept for each individual
is set by default and does not need to be passed
here.
But when I run the code without subject ID as an explicit random effect, subject ID is not listed as a random effect. For a sanity check I ran it once with subject ID as random effect and once without (with the assumption it does at default) to compare and I did get slightly different results… I don’t want to be in a situation where I am accidentally adding subject ID as a random effect twice?
TLDR; when running linear mixed effects model with q2 longitudinal, after running first distance, does one explicitly state subject ID as a random effect (like how you would do when running lme in R) or is this already accounted for within this model in qiime2?
many thanks for your help 
Zoë