Related to #6
Now that we are going to have multiple versions of file products after running different software, we need a way to encode this information in file names. Particularly databases (eg chocophlan database used for metaphlan).
This is complicated by a couple of things. For example, for Humann outputs, do we also need to encode the metaphlan database used for the input taxonomic profiles? For both metaphlan and humann, do we need to include the human genome version used in kneaddata?
We should probably be tracking provenance in airtable, and mostly use file-name changes to avoid clobbering files.
Related to #6
Now that we are going to have multiple versions of file products after running different software, we need a way to encode this information in file names. Particularly databases (eg chocophlan database used for metaphlan).
This is complicated by a couple of things. For example, for Humann outputs, do we also need to encode the metaphlan database used for the input taxonomic profiles? For both metaphlan and humann, do we need to include the human genome version used in kneaddata?
We should probably be tracking provenance in airtable, and mostly use file-name changes to avoid clobbering files.