alpha and beta output problems

Hello everyone.

I hope you are doing great.

I have three questions and I would really appreciate it if anyone can help, please.

First part:

I have 48 samples which are gathered from two farms (28 sample from one farm (MOL) and 20 sample from another farm (PIC))

Second part

Again, as you can see qiime output is saying that I have two equal numbers of sample (MOL (n=20) PIC (n=20)) which is wrong. Why is that?

Third part:

After performing alpha diversity on the dataset I have only received shannon and observed_otus indexes but I don’t have others like chao1. Why is that?

The command lines are the following:

$ qiime diversity core-metrics-phylogenetic
--i-phylogeny rooted-tree.qza
--i-table table.qza
--p-sampling-depth 10000
--m-metadata metadata.tsv
--output-dir core_metrics_results

If other information is needed for clarification I would be happy to provide.

Thank you,


HI @ari_sh70,

Probably because you are choosing a sampling depth that is higher than the sampling depth of the 8 missing samples:

Use qiime feature-table summarize --i-table table.qza to see the sampling depth of all samples, to determine an appropriate depth.

Because the pipeline you are running does not output chao1, you need a different method. See qiime diversity alpha to choose individual alpha diversity metrics you wish to estimate, that will have an option for estimating chao1

Warning: note that chao1 is probably not a valid metric, depending on how you processed your data. If I recall correctly, chao1 uses singletons in the calculation. The dada2 and deblur denoising methods filter out singletons by default, so the outputs cannot be used with chao1. If you used those methods or any method to filter out singletons, then you should not use chao1 or other diversity metrics that incorporate singletons.


Thank you very much @Nicholas_Bokulich for your answer. It worked. So the least number was 7710 and I set this number for --p-sampling-depth and now I have all the 48 samples.

Thank you

Now I have 48 samples, and the beta diversity graph is showing that there is a significance between two farms (blue and red dots) but on the other hand the p-values is saying that they are not significant. What could be the scientific reason behind that?

Beta diversity graph


Thank you all

Looks like you are comparing alpha and beta diversity, which is a bit like :apple:s and :tangerine:s. Use the beta-group-significance action to run a PERMANOVA test, that’s a better test to compare against the PCoA.


Thank you @Nicholas_Bokulich Yes. I have done that comparison.

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