Hi everyone,

I am analyzing 16S microbiome data to compare certain OTUs between two treatments, but I am not sure if my analysis pipeline is legit:

I have two samples (1 and 2) from each treatment (A and B).

1. First, I am using only OTUs that appear both in A1 and A2.
2. I am extracting those OTUs from B1 and B2 (if exist).
3. To deal with different read depths, I rarefy the samples to the one with the lowest depth.
4. I am merging the samples by treatment.
5. Transforming reads to relative abundance.
6. Since the difference might be really small (i.e, 0.000125 to 0.002) I am transforming to log10 and then using R's rescale function to fit the numbers again between 0 to 1.

Finally, I am plotting the heatmap so I can see a nice difference in colors between certain OTUs in low numbers.

However, I am worried about step 6. I there a better way to deal with this data, but still maintaining high sensitivity for the small numbers so the differences can be clearly visualized in the heatmap?

Thanks