Metadata error in retrieving column

Hello, I am not sure why I keep getting the error message: "Invalid value for '--m-metadata-file': There was an issue with retrieving column 'sample-id' from the metadata"

I will like to split the feature table based on the sample-ids such that values
in that column are evenly represented across training and test sets

Code:
%%bash
qiime sample-classifier split-table
--i-table ~/miniconda3/sediment-non-saline.qza
--m-metadata-file ~/miniconda3/metadatasns.tsv
--m-metadata-column sample-id
--o-training-table ~/miniconda3/split-table
--o-test-table ~/miniconda3/split-table
--o-training-targets ~/miniconda3/split-table
--o-test-targets ~/miniconda3/split-table

The files I use are attached.

metadatasns.tsv (8.7 KB)
sediment-non-saline.qza (4.6 MB)

I think the issue is that sample-id is basically the pandas index, and not a column. So, as this is sample-classifier you need to add another column, like group, that specifies which groups the samples belong to.

sample-id group
sample1 case
sample2 control
... ...
1 Like

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