Open kzuberi opened 9 years ago
The attribute processing scripts are expecting two column (attribute_id, attribute_description) files but the input is actually a triplet of (attribute_id, attribute_name, attribute_description). This causes a join during processing to produce an empty set of descriptions.
I've committed a fix, but need to update the documentation. Needs testing.
@haroldr, let me know if this works out, thanks.
Only attribute ids such as IPR016132 for InterPro appear to be making it as far as generic_db.