StructureSimilarity "Matrix don't have the same number of points" between two matched files

[DarioMarzella]

Describe the bug When trying to run compute_lrmsd_fast() on two files (model and target structure), the following error is raised: ValueError: ("Matrix don't have the same number of points", (524, 3), (528, 3))

Environment:

• OS system: Linux
• Version: CentOS Linux 7 (Core)
• Branch commit ID: 7a6d7cd
• Inputs: python run.py (in the attached folder)

To Reproduce Steps/commands to reproduce the behaviour:

1. open the attached folder
2. run python run.py

Expected Results It should compute the l-RMSD

Actual Results or Error Info it raises the following error: ValueError: ("Matrix don't have the same number of points", (524, 3), (528, 3))

Additional Context The model and ref file have been matched, thus only the common residues are present.

Attached folder: pdb2sql_issue_58.zip

[CunliangGeng]

The cause of the error is that ref.pdb have double occupancies, e.g.

ATOM    779  N  ATHR M  94     -21.816   9.229  30.233  0.50 38.91           N  
ATOM    780  N  BTHR M  94     -21.729   9.244  30.235  0.50 38.76           N  
ATOM    781  CA ATHR M  94     -21.296   8.442  29.129  0.50 38.41           C  
ATOM    782  CA BTHR M  94     -21.316   8.480  29.067  0.50 38.04           C  

Use the pdb tool pdb_delocc.py to preprocess ref.pdb, then the rmsd calculation should work well.

BTW, in the run.py, the lrmsd = sim.compute_lrmsd_fast(method='svd', name = atoms) does not accept name as a parameter, you should remove it.

[NicoRenaud]

well it does accept name on issue55 branch, That we should merge now :)

[DarioMarzella]

Selecting only one occupancy with https://github.com/LilySnow/PDB_related/pdb_selalt.py solved the problem. Thanks.