plinder-org / plinder

Protein Ligand INteraction Dataset and Evaluation Resource
https://plinder.sh
Apache License 2.0
166 stars 9 forks source link

Add functionality to deleak specific test set #16

Closed yusuf1759 closed 3 weeks ago

yusuf1759 commented 3 months ago
github-actions[bot] commented 3 months ago

Coverage report

Click to see where and how coverage changed

FileStatementsMissingCoverageCoverage
(new stmts)
Lines missing
  src/plinder/core
  __init__.py
  src/plinder/data
  __init__.py
  get_system_annotations.py
  splits.py 131, 161-195, 290, 304-307, 704
  src/plinder/data/pipeline
  config.py
  io.py 147-164, 172-181
  utils.py 420-423, 450
  src/plinder/data/utils/annotations
  aggregate_annotations.py 1154-1155
  get_ligand_validation.py
  interaction_utils.py
  ligand_utils.py 333, 429
  rdkit_utils.py
  save_utils.py
  src/plinder/eval/docking
  utils.py
Project Total  

This report was generated by python-coverage-comment-action

OleinikovasV commented 3 months ago

The script here is very specific for PoseBusters format for deleaking - using format "{PDBID}{CCD_CODE}" with uppercase and only single atom. This would not work for the general case, eg. specifically chosen cases from PLINDER - using PLINDER system_id. I would recommend moving this specific example tailored code to the notebook, and keep split.py as general code.