Esmecata

Modify

* use new UniProt API.
* remove beta option and old UniProt queries.

Fix

* issue with new URL for UniProt mapping.
* issue in test with new UniProt release.

Fix

* issue with time out in `proteomes.py`.

Add

* `--minimal-nb-proteomes`: Choose the minimal number of proteomes to be selected by EsMeCaTa, if a taxon has less proteomes, it will be ignored (issue 11).
* add a jupyter notebook explaining EsMeCaTa method.

Fix

* issue when parsing SPARQL query result. The filtering of proteomes with BUSCO score in SPARQL query was incorrect.

Refactor the code in EsMeCaTa to make it clearer.

Add

* New functions with the refactoring.
* Comments have been added to make the code more understandable.
* Singularity recipe.
* Timeout in query, this should help for issue 7.
* Tests for the new functions.
* Logging to show log and creates a log file at each run with the command line.

Fix

* Decrease the number of proteins given to Uniprot in a query, to avoid issue 7 (from 20,000 to 15,000).
* Fix issue in sub sampling with incorrect threshold.
* Fix numerous typos.

Modify

* Update esmecata workflow picture and readme.

Add

* EsMeCaTa step `workflow` (performing steps `proteomes`, `clustering` and `annotation`).
* Time measures in metadata jsons.
* Article data.
* Workflow test using conda (for MMseqs2) and GitHub Actions.
* `--beta` should work to be used with new UniProt REST API (still in development).

Fix

* Issue in statistics computation (wrong number return).