Error: Plugin error from diversity “All numbers are identical in kruskal“

[ilhem (sa)]

Hello ,

I’m trying to visualize alpha diversity from core metrics of goods average and I got an error “All numbers are identical in kruskal“ =. I attached the error file here with the command that I run

.

I run the visualization of other alpha diversity metrics ( Shannon Chao1 ..) with the same metadata and table .qza and all run ok .

Thank you in advance for any help !

[thermokarst]

Hi @ilhem, can you either attach the log file mentioned on the last line of your screenshot, or, rerun with the --verbose flag, and provide the complete error message? Thanks!

[ilhem (sa)]

Hi @thermokarst
Thank you for your reply, here a screenshot of the error with the --verbose flag

[thermokarst]

Hi @ilhem!

I think what this error message is saying is that at least one of your metadata columns is producing groups where the values are all identical (but, I’m not 100% sure of that assessment). Would you be able to send me (in a private message) links to download GEotable.qza, goods_coverage_vector.qza, and metadata? I would like to reproduce this issue locally. Thanks!

[thermokarst]

Hi @ilhem, thanks for sharing your data with me, that was really helpful!

If you run the following, the problem be more apparent:

qiime metadata tabulate \
  --m-input-file goods_coverage_vector.qza \
  --o-visualization goods_coverage_vector.qzv

Every one of your samples has a Good’s Coverage value of 1, so all of the samples are identical, in terms of that measure of alpha diversity. Taking a step back - how was this feature table produced? Since Good’s coverage requires the presence of singletons, I think it would be a mistake to try and calculate this measure on a table produced by a method that removes singletons (or doesn’t generally produce singletons, like DADA2), but I am curious to hear other opinions on the matter.

Good luck!

[ilhem (sa)]

Hi @thermokarst,
thank you for your reply I appreciate your advice on this!

how was this feature table produced?

this feature table was sent by another partner as a txt file so I used the command of qiime tool import to convert it to qza and I don't know what method they used to generate it.