gabefoley / asr_curation Goto Github PK

Need to create a default setting that allows all of the data to be processed.

Users should be allowed to define custom annotations. Use the user-defined custom annotations to fetch, visualise and create annotation files ( for Figtree viewing).

update KARI Class names

Most people using the asr curation pipeline for their protein family would have custom annotations after the alignment is complete. This setting would allow them to use their own scripts for annotating.

Add files for which version does not need to be tracked.

Need to start creating pytest tests and to set them to run on merge

KeyError 'strands' is thrown and fails hard if the secondary structure dictionary doesn't have this key in it.

Need to update the custom annotation columns in the example config.

And check that the annotations text file (for FigTree viewing) is working correctly.

Fix uniprot_annotations_pagination script -
fix the import statement
fix the lineage columns
fix not to include empty uniprot ids.

Fix validate_ids.py script-
not include rows if there is no information.

Fix create_annotation_file.py-
fix with new uniprot column names

Fix seqcurate.py-
remove print statement

Fix rules in snakemake file-
change the rule create annotations to create ids and similar change with script names in the rules.

Let the pipeline run and make trees even if it fails validation or doesn't have UniProt sequences.

Update image links in README to match new folder structure

Need to update the code in the async calls to UniProt -

with async_timeout.timeout(timeout):

Can be recreated with the full KARI dataset.

This line
annot_df['lineage_forma'] = zip(*annot_df['lineage'].map(split_lineage))

Seems to fail, assumably when there is no info in the cell.

There is a new uniprot api and the scripts needs to be updated with the new uniprot api.

The chosen data, subsets, and columns that get annotated don't demonstrate any use case or story.

KARI custom annotations script does not work after change in UNIPROT API names.

The custom annotations script will be split into 2 scripts - generic one and the custom one. generic one will be common for all workflows.

Currently the pipeline looks for an EC number to be added into the filename in order to query BRENDA for information on that EC number, but we can (and should) get this information from the UniProt column ec.

Need to make sure we can handle multiple ec numbers as well.