DeepRank / pdb2sql

Fast and versatile biomolecular structure PDB file parser using SQL queries
https://pdb2sql.readthedocs.io
Apache License 2.0
24 stars 12 forks source link

StructureSimilarity.compute_lrmsd_fast() and StructureSimilarity.compute_lrmsd_fast() provide different results #57

Closed DarioMarzella closed 2 years ago

DarioMarzella commented 4 years ago

Describe the bug In a case where the model and the reference file cannot be easily aligned, StructureSimilarity.compute_lrmsd_fast() and StructureSimilarity.comp

ute_lrmsd_fast() provide different results (respectively, 0.837 and 3.38)

Environment:

To Reproduce Steps/commands to reproduce the behaviour:

  1. open the attached folder named different_lrmsd
  2. run python get_rmsd.py

Expected Results StructureSimilarity.compute_lrmsd_fast() and StructureSimilarity.compute_lrmsd_pdb2sql() should be providing the same output CA l-RMSD.

Actual Results or Error Info The two methods provide very different CA l-RMSDs (respectively, 0.837 and 3.38)

Additional Context A .pse PyMol session is included in the attached folder. It is possible to see how the _model is technically very similar to the ref file, but they are not properly aligned by PyMol. StructureSimilarity.compute_lrmsd_pdb2sql() as well is probably failing in aligning them, but StructureSimilarity.compute_lrmsd_fast() seems to work properly.

different_lrmsd.zip

DTRademaker commented 2 years ago

Dear Dario,

I could not reproduce your problem because of an error "Residue numbering not identical in ref and decoy". After renumbering the 'ref.pdb' so that it had the same numbering as 'model.pdb' (see atachement) it gave very similar results for the two similarity methods: "l-RMSD fast: 3.414 l-RMSD pdb2sql: 3.38". So renumbering the pdbs seems to solve your problem, but maybe also update your pdb2sql library or just try to restart your pc? (based on that you did not get the same error as I did)

hope it helps, Daniel different_lrmsd.zip

github-actions[bot] commented 2 years ago

This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.

github-actions[bot] commented 2 years ago

This issue was closed because it has been inactive for 7 days since being marked as stale.