Does QIIME2 have an equivalent of beta_significance.py from qiime1?

Hi all,

I’m looking for an equivalent to beta-significance.py from qiime1?

Does such a plugin exist? I’ve combed through the tutorials and while diversity supports group significance, I’m looking to compare sample significance.

Thanks!

Hi @William_G_Ryan! Have you had a chance to look at the docs? In particular, this section of the Moving Pictures Tutorial, and, the docs for q2-diversity.

:qiime2: :t_rex:

My apologies, I accidentally typed the wrong plugin name I’m referring to beta_significance.py from qiime1.

Hey @William_G_Ryan, no difference, the same two links I sent have you covered. In fact, the MP Tutorial has a worked example of beta group significance. :open_book:

I understand. However, I’m not looking to compare beta group significance. I’m looking to compare between samples, not groups of samples. If you look at beta_significance.py, it requires an OTU table and tree and does pairwise comparisons between each sample and generates a p-value for each comparison with unifrac monte carlo simulations.

Beta group significance requires a group category parameter. If I attempt to use beta group significance with my sample names, it fails because they all unique samples names.

Hi @William_G_Ryan,

The short answer is that there isn’t a re-implementation in QIIME 2 for determining if pairwise beta diversity calculations are significant against some null hypothesis.

I think, in general, this is because there’s been movement toward larger studies that make per sample pairwise comparisons somewhat moot in the face of larger, cross sectional data. At this point, I would discourage this approach to the data. If you need a specific test for paired samples in group, the pairwise sample tutorial from the longitudinal plugin may be helpful.

I think the best advice is to work with group-level rather than per-sample comparisons.

3 Likes

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