search

Electrostatics / pdb2pqr

PDB2PQR - determining titration states, adding missing atoms, and assigning charges/radii to biomolecules.
http://www.poissonboltzmann.org/
Other
122 stars 34 forks source link

Failing with over 99999 atoms. #288

Closed jacob-r-anderson closed 2 years ago

jacob-r-anderson commented 2 years ago

pdb2pqr 3.4.1, Ubuntu 18.04.

I have a large protein I am hoping to input to pdb2pqr30. Oddly, this works on an older version of pdb2pqr that was packaged with pymol (--version doesn't give me a version number with my pymol version). When I run with pdb2pqr30 I get this error for the same input file.

ValueError: invalid literal for int() with base 10: 'A0000'

If I check the file for lines containing this 'A0000' I find this as the first hit.

ATOM  99998  CA  ASP G1096     134.223 148.670 157.858  1.00 46.94           C
ATOM  99999  C   ASP G1096     135.627 148.143 158.120  1.00 46.94           C
ATOM  A0000  O   ASP G1096     136.582 148.529 157.436  1.00 46.94           O
ATOM  A0001  CB  ASP G1096     133.541 147.984 156.665  1.00 46.94           C

I presume this is due to a problem handling the letter A in the in the second column.

sobolevnrm commented 2 years ago

Can you please provide a PDB file (you will need to add .txt suffix or compress as ZIP) so we can reproduce the error?

intendo commented 2 years ago

According to https://www.wwpdb.org/documentation/file-format-content/format33/sect9.html:

A model cannot have more than 99,999 atoms. Where the entry does not contain an ensemble of models, then the entry cannot have more than 99,999 atoms. Entries that go beyond this atom limit must be split into multiple entries, each containing no more than the limits specified above.

sobolevnrm commented 2 years ago

@intendo has a good point. @jacob-r-anderson do you have this structure in CIF format?

jacob-r-anderson commented 2 years ago

@intendo @sobolevnrm PDBs with over 99,999 atoms are the primary data object we have worked with for several years. One is attached. Other softwares handle these PDBs (Coot, Chimera, ChimeraX, Phenix).

Microtubule.pdb.txt.zip

intendo commented 2 years ago

@sobolevnrm should we create a new issue (feature request) to handle PDB files that don't conform to the PDB 3.3 format? It would require a change to the parser and then propagate throughout the code base from there.

sobolevnrm commented 2 years ago

@sobolevnrm should we create a new issue (feature request) to handle PDB files that don't conform to the PDB 3.3 format? It would require a change to the parser and then propagate throughout the code base from there.

Yes, I'll create a low-priority issue for this. In the meantime, @jacob-r-anderson can you try CIF format?