non GPCR Proteins without cysteine bridges cannot be parsed

[antilotus]

Hi,

I am interested in using this tool for the transmembrane odor receptor I am studying. However as my structure does not have any cysteines bridges I only get this error message when I try to run it:
"""
IndexError Traceback (most recent call last)
File ~/hdd/mdrun/Gromacs/12percholtip3p/SpicOR29/system_SpicOR29_ORCO/alloviztest/Spicalloviz.py:3
1 import AlloViz
----> 3 system = AlloViz.Protein(pdb="/home/albert/hdd/mdrun/Gromacs/12percholtip3p/SpicOR29/system_SpicOR29_ORCO/SpicOR29_ORCO.pdb",
4 trajs="/home/albert/hdd/mdrun/Gromacs/12percholtip3p/SpicOR29/system_SpicOR29_ORCO/Out/Production_MD_r1/md_fit.xtc",
5 path="/home/albert/hdd/mdrun/Gromacs/12percholtip3p/SpicOR29/system_SpicOR29_ORCO/alloviztest")
6 #system.calculate(pkgs="correlationplus_CA_Pear", cores=12)
7 system.calculate(pkgs="pytraj_CA", cores=30)

File ~/micromamba/envs/alloviz/lib/python3.9/site-packages/AlloViz/AlloViz/Classes.py:244, in Protein.init(self, pdb, trajs, GPCR, name, path, protein_sel, **kwargs)
241 """:class:~MDAnalysis.core.universe.Universe"""
243 # Bonded cysteines must be identified to remove them from non-covalent contacts calculations (i.e., PyInteraph2_Contacts)
--> 244 self._bonded_cys = trajutils.get_bonded_cys(self._pdbf)
246 # _dihedral_residx returns a list of indices of all the protein's residues in the Universes that are not in the extremes of the chain(s)
247 # These will be the residues for which Dihedral correlations can be calculated (residues in the extremes have some dihedrals missing)
248 # MDEntropy_AlphaAngle uses information of residues i-1, i, i+1, i+2; so end=-2 is passed to also exclude the second-to-last residue of the chain(s)
249 _res_arrays = np.split(
250 self.protein.residues.resindices,
251 np.where(np.diff(self.protein.residues.resnums) != 1)[0] + 1,
252 )

File ~/micromamba/envs/alloviz/lib/python3.9/site-packages/AlloViz/AlloViz/trajutils.py:376, in get_bonded_cys(pdbf)
361 r"""Identify disulfide bond-forming cysteines' sulphur atoms
362
363 Returns a list of the residue indices of cysteines whose sulphur atoms form a
(...)
373 Filename of the already-processed protein PDB structure to use for detection.
374 """
375 bonded_cys = []
--> 376 pdb = parmed.read_pdb(pdbf)
377 mask = parmed.amber.mask.AmberMask(pdb, ":CY?@sg")
378 for sel in mask.Selected():

File ~/micromamba/envs/alloviz/lib/python3.9/site-packages/parmed/formats/pdb.py:447, in PDBFile.parse(cls, filename, skip_bonds, expanded_residue_template_match, all_residue_template_match)
445 _map_bond_types_from_templates(inst.struct, *residue_libraries)
446 else:
--> 447 inst.struct.assign_bonds()
449 inst._postprocess_metadata()
450 inst.struct.unchange()

File ~/micromamba/envs/alloviz/lib/python3.9/site-packages/parmed/structure.py:886, in Structure.assign_bonds(self, *reslibs)
884 cysteine_sg = set()
885 for res in self.residues:
--> 886 templ = _res_in_templlib(res, all_residues)
887 if templ is None:
888 # Don't have a template for this residue. Keep a note and go on
889 unassigned_atoms.update(res.atoms)

File ~/micromamba/envs/alloviz/lib/python3.9/site-packages/parmed/structure.py:4311, in _res_in_templlib(res, lib)
4309 if len(res.name) == 3 and residue.AminoAcidResidue.has(res.name):
4310 return lib[residue.AminoAcidResidue.get(res.name).abbr]
-> 4311 if residue.DNAResidue.has(res.name):
4312 return lib[residue.DNAResidue.get(res.name).abbr]
4313 if residue.RNAResidue.has(res.name) and residue.RNAResidue.get(res.name).abbr != 'T':

File ~/micromamba/envs/alloviz/lib/python3.9/site-packages/parmed/residue.py:51, in BiomolecularResidue.has(cls, thing)
49 return thing in cls.all_residues
50 try:
---> 51 cls.get(thing)
52 except KeyError:
53 return False

File ~/micromamba/envs/alloviz/lib/python3.9/site-packages/parmed/residue.py:216, in DNAResidue.get(cls, key)
196 """
197 Gets the nucleic acid corresponding to either the residue name or
198 abbreviation. It is case-insensitive.
(...)
213 an DNA residue.
214 """
215 try:
--> 216 if key[-1] in '35':
217 return cls._all_residues_by_abbr[key[:-1].upper()]
218 return cls._all_residues_by_abbr[key.upper()]

IndexError: string index out of range
"""

[frannerin]

Thank you for using AlloViz! It seems that the issue stems from a non-standard residue in your structure, does your protein have any non-standard residue names or non-protein molecules?

For non-standard residue names, please try to pass the special_res argument to AlloViz.Protein mapping the 3-letter to 1-letter residue codes, e.g.: special_res = {"HSE": "H"}

To select only protein molecules from the system, please adjust the protein_sel argument of AlloViz.Protein with an appropriate MDAnalysis selection, by default protein_sel is defined as: '(same segid as protein) and (not segid LIG) and (not chainid L)'