Rolf Huehne wrote: >Bob Hanson wrote: > > >>OK, well, then, since they messed up there, I highly recommend you use >>"connect" just after loading. I've added the test for between-model >>bonding in Jmol 11.3.48 >> >> >> >I don't fully trust the connect mechanism in Jmol (I saw for example >hydrogen atoms bonded covalently with 2 other atoms in close proximity >[a single model file]). Therefore I would prefer to display the >"official" bonds. > > > specific example, please? It's true, that if two atoms are very close Jmol doesn't apply valence rules to only bond H to one of them. After all, there are plenty of situations where H is bonded to 2 atoms (boranes, for example).
BUT, that said, Jmol 11.3.51 will follow the PDB rules and apply a single set of connections to ALL models in the set. So you should have no problem using set autobond false load ...... I could imagine that this may cause grief with people who misappropriate the PDB multimodel format, so I've made it so that one COULD have multiple CONECT files, and these are flushed each time a new MODEL command is found. Doesn't sound like that is any issue with true PDB files. >The PDB format documentation at >"http://www.wwpdb.org/documentation/format23/sect10.html"; tells for >multiple model cases the following: > >"For NMR entries, CONECT records for all models are generated describing >heterogen connectivity and others for LINK records." > >This is not really clear. But it becomes more clear in combination with >the documentation for "LINK" records at >"http://www.wwpdb.org/documentation/format23/sect6.html#LINK": > > Jmol does not currently read LINK records. I believe that is correct, because: LINK *Overview* The LINK records specify connectivity between residues that is not implied by the primary structure. Connectivity is expressed in terms of the atom names. This record supplements information given in CONECT records and is provided here for convenience in searching. That implies to me that it is not just a supplement, but a redundant duplication "for convenience" only. Right? >"For NMR entries only one set (or model) of LINK records will be supplied." > >This means that a single set of "CONECT" records has to be applied to >all models (separately). > >I analyzed the usage of "CONECT" records within all asymmetric and >biological unit PDB format files of the Protein Data Bank. The outcome >can be summarized as follows: > >1) There are never multiple sets of "CONECT" records. > (confirming the interpretation of the documentation) > >2) Statistics > asymmetric, with conect: 36791 > asymmetric, with conect, with missing atoms: 40 > asymmetric, with conect, > 1 model: 2336 > asymmetric, with conect, > 1 model, with missing_atoms: 38 > biological, with conect: 59933 > biological, with conect, with missing atoms: 3624 > biological, with conect, > 1 model: 8126 > biological, with conect, > 1 model, with missing_atoms: 885 > >The full results are (temporarily) available either as compressed CSV >format file (692 kb) or as compressed MySQL dump (711 kb): > >http://www.fli-leibniz.de/~rhuehne/jmol/analyze_conect-2007_11_25-csv.zip > > This is very helpful. Thank you! >http://www.fli-leibniz.de/~rhuehne/jmol/analyze_conect-2007_11_25-mysql.zip > >The colums are defined as follows: >1) id: PDB ID >2) unit_type: asymmetric/biological >3) unit_number: biological unit number or zero for asymmetric unit >4) maximum_model_number: largest model number found, -1 if no "MODEL" >line found >4) models_with_conect_count: number of "CONECT" record sets >6) connected_atom_count: number of atoms referred to in "CONECT" records > (missing atoms were not counted because they couldn't be classified) >7) connected_hetatom_count: number of hetatoms referred to in "CONECT" >records (missing atoms were not counted because they couldn't be >classified) >8) missing_atom_count: number of atoms/hetatoms referred to in "CONECT" >records that were not found in at least 1 model >9) missing_atoms: complete list of missing atoms in Jmol-like manner > (atom number/model_number) > >Since Jmol currently applies the "CONECT" record only to the last model >before the "CONECT" record it doesn't comply with the PDB format >specification. >I think this should be changed. > > > Jmol 11.3.51 a) applies a single final CONECT record set to all models in a PDB file. b) ignores any bonds to atoms that do not exist in a given model. c) allows multiple CONECT records, one per model, if present. This is not standard PDB format, of course. Note that in general this is only applicable when one has set autobond OFF prior to file loading, because generally Jmol makes its own connections. Bob >Regards, >Rolf > >------------------------------------------------------------------------- >SF.Net email is sponsored by: The Future of Linux Business White Paper >from Novell. From the desktop to the data center, Linux is going >mainstream. Let it simplify your IT future. >http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4 >_______________________________________________ >Jmol-users mailing list >Jmol-users@lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/jmol-users > > -- Robert M. Hanson Professor of Chemistry St. Olaf College Northfield, MN http://www.stolaf.edu/people/hansonr If nature does not answer first what we want, it is better to take what answer we get. -- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900 ------------------------------------------------------------------------- SF.Net email is sponsored by: The Future of Linux Business White Paper from Novell. From the desktop to the data center, Linux is going mainstream. Let it simplify your IT future. http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4 _______________________________________________ Jmol-users mailing list Jmol-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jmol-users
