Dear Charles<br/><br/>I update my program to 2.47 version, the error is solved. Thanks very much.<br/><br/>Best regards<br/>Jialiang Chen 在 2017-12-31 21:29:05，Charles Schwieters <[email protected]> 写道： > >Hi 陈家良-- >> > >> >> >> >> My protein has a Mg2+ in the pdb file, but the psf file generate by >> >> "pdb2psf" >> >> command did not contain the Mg2+ information. >> > >> > >> >In the PDB file, if you change the HETATM literal for to "ATOM ", >> >such that the beginning of the associated line reads >> > >> >ATOM XXXX MG+2 MG2 >> > >> >pdb2psf should produce a working PSF file, and protocol.initCoords >> >should read the coordinates in a script, if preceded by >> >protocol.initStruct. Alternately, you could simply use >> >protocol.loadPDB, and then there would be no PSF file required. >> > >> >I tested this with Xplor-NIH version 2.47. Manually editing PSF files >> >is error-prone and not recommended. >> > >> > >> Replacement of "HETAM" with "ATOM" in PDB file indeed generated the >> psf file containing Mg2+ informnation. But the script reported >> another error "PDBTool:read: atom not found in structure: " by >> protocol.initCoords reading the coordinates (all atoms in the pdb >> file were not found). It seems that the psf file was still not >> working. > >protocol.initCoords will report some messages, but generally properly >match atoms after PDBTool is invoked. You can see which atoms' >coordinates are not read with this snippet: > >for atom in AtomSel("not known"): > print atom.string() > > >> The protocol.loadPDB worked well. But when I input >> coordinates of GDP ligand and some axis by protocol.initCoords (psf >> file also loaded by protocol.initStruct), an error is "SystemError: >> xplor-nih error: Error: named selection with name pseudo contains an >> inconsistent number of atoms.". I also tried loaded coordinates of >> GDP and axis by protocol.loadPDB, it also did not work. <br/>My >> program version was 2.35, is it ok? > >2.35 s quite old. If this error occurs with version 2.47, I'd like to >investigate further. > >best regards-- >Charles
_______________________________________________ Xplor-nih mailing list [email protected] https://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
