Comparing ASV counts between groups of samples

[georgia (Georgia Parks)]

Hello! I have an issue that I have been working on for many months now and cannot find a consensus on how to approach it.

After working with my data in QIIME2 and exporting the feature table with taxonomy, I am wanting to statistically test the abundance of one specific genera (at level 6) between groups of samples.
For example, there are 9 samples in group A and 10 samples in group B. I am wanting to use the number of ASV “hits?” for the genus Halomonas in each group of samples to test if there is any significant difference between the groups. So far, I have used the Welch’s student t-test. Is there another statistic test that can handle a small amount of samples, many zeros, and unequal sample sizes?

Thanks for any help!

[Nicholas_Bokulich]

Hi @georgia,

@georgia

After working with my data in QIIME2 and exporting the feature table with taxonomy, I am wanting to statistically test the abundance of one specific genera (at level 6) between groups of samples

Sounds like instead of exporting you should just be using ANCOM inside QIIME2. It is designed for this type of data.

As far as I know, Welch's t-test is not suited for this type of data — compositional, non-normal distributions.

ANCOM will test ALL taxa, not just that one genus (unless if you filter your feature table) but maybe that's not a bad thing

I hope that helps!