Plugin error from diversity must provide at least one column bioenv

Hi @ibrahim_mekhimar,
As the error indicates your metadata file needs to have at least one column (excluding sample-ids) for the visualization to work with. For example a grouping variable.

1 Like

Many thanks @Mehrbod_Estaki for your reply. I have modified my metadata to the following form but it resulted in the same error.

Hi @ibrahim_mekhimar,
Have you validated your metadata file using Keemei by chance? Otherwise I can’t really think of anything else from the top of my head, would you be willing to share your distance matrix and metadata file? You can send this via DM if you rather not post it here.

thanks for your time @Mehrbod_Estaki ,
yes i have validated it using keemei and it was valid. i sent you distance matrix and metadata file.

1 Like

Hi @ibrahim_mekhimar,
Thanks for providing your data. I think I found the problem. From the bioenv help page:

Usage: qiime diversity bioenv [OPTIONS]

  Find the subsets of variables in metadata whose Euclidean distances are
  maximally rank-correlated with distance matrix. All numeric variables in
  metadata will be considered, and samples which are missing data will be
  dropped. The output visualization will indicate how many samples were
  dropped due to missing data, if any were dropped.

Notice how the plugin is attempting to work with numeric variables. Your metadata file only has one numeric category (Year) and in that one category there is only 1 number so there is nothing to run correlation with. If I arbitrarily change those values to different numbers your command works fine. I suspect you may be be confusing what this plugin does vs what you want it to do? :thinking:

1 Like

Hi @Mehrbod_Estaki many thanks for your time agian and sorry for my bad questions.
I think that i am confused about plugins correlated to feature-table.
My samples divided into four groups as you see in the metadata file and i want to drive the analysis to compare the four groups but i don’t know how to use feature-table group .
Should i use a grouped_table as input for collapsed_table or vice versa.
Sorry for bothering you again

1 Like

Hi @ibrahim_mekhimar,
No bother at all! It is a bit hard to recommend what you should or shouldn’t without more information about your experiment design. And since this is drifting away from the original topic could you please start a new topic and in that give us a bit more explanation as to what it is exactly you are trying to do and compare. Things for example, what type of analysis you are trying to do, exploratory analysis (PCoA), group comparisons like alpha, or beta diversity? Are you trying to discover individual taxa different between groups? Are you working with multiple variables? Continuous, categorical? Just some of the basics like this will help us navigate you better.
That being said, I would also advise going through the Qiime2 tutorials as well, there are lots of great examples of all these through them. For example, an overview of the methods available are nicely drawn out here and a step by step example of basic workflow is found here.