Hello! I have my ASV relative abundance data (normalized per sample), and I'm visualizing the taxonomic classifications across the various months I sampled. I'm working in R, and I realized that I haven't accounted for the uneven sampling depth per month (different months have a different number of samples). Is there a way I could account for this? Right now, I'm using the normalized abundances to calculate monthly Phyla abundance and using that for a stacked barplot.
Hello Mahima,
EDIT: I mistakenly though you were asking about sampling depth during sequencing, and not balance/unbalanced number of samples in each cohort!
I've been able to publish imperfectly balanced cohorts, as long as we had enough reps. PMC5319707 CD n=49, UC n=60, HC = 9
PMC6775999 n = 20 samples taken, got n=4 at the first timepoint due to low-biomass 
If you want to prove to your self (or reviewer three!) that you have enough reps, consider a power analysis as described in PMC10297957
Normalizing sampling depth been much debated over the years.
I recommend this paper:
"repeatedly rarefying" == rarefaction == bootstrapping
I recommend this plugin, which implements bootstrapping for use with Qiime2!
3 Likes
Thank you so much!! I appreciate it!