Can we do only two numbers of samples for microbial diversity, taxonomic and ANCOM analysis

Hello! Can I please get clarification?

Can we do only two samples - for 16s rRNA analysis - for studying the microbial abundance, diversity (alpha and beta), ANCOM and taxonomic composition

I know we need minimum 5 in a group and we need 2 groups at least to perform statistical analysis. But my question is can we do only 2 samples and can get any results?

If you meant technically whether QIIME2 would allow you to do it, then yes! As long as you have 2 samples (otherwise there is no “distance”) QIIME would allow processing of diversity and abundance etc. For stuff like taxonomic composition, even 1 sample is not a problem (but of course you need to import a database to train the classifier).

Hi @srini,

To build on @Focussash’s answer, you can run some basic commands (denoising, taxonomic clasification) and distance calculations. Whether or not these constitute “results” is up to you. I’m not sure if core diversity metrics will work because I haven’t tried PCoA with 2 samples. You won’t be able to get more than one axis so i would guess not. You cannot run statistical tests.

Best,
Justine

1 Like

Hello, my first time here… I have a case like this, I have to do an analysis of a biofilter and I only have two samples (two different times), I already did all the analysis with qiime2 and I have the taxonomic assignment, and I can see the barplots of the two samples. However, I would like to know if it is possible to do some analysis of diversity or abundance between the two samples, to see statistical significance.
Thanks in advance
Cony

Hi @conscc,

Welcome to the :qiime2: forum!

Unfortunately, it's not posssible to do much with two samples. Statistics need a distribution of multiple samples; most of the tests in QIIME need at least 5 samples per group to meet their basic assumptions, and in most casess Ive see (with the caveat that I work in cross sectional human data), you need many more to get interpretable results.

Best,
Justine

Thanks for the reply!! :pensive:

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.