David Leader wrote: >I asked about representing mainchain-sidechain H-bonds for short >three-dimensional hydrogen bonded motifs. Bob replied: > > > >>If the motifs are well-described in the PDB file, this would be no >>problem. If they are to be generated from scratch, that's more of a problem. >> >> > >No. The way the thing works is that the H-bonds are generated from >the PDB files using HBPlus, which is then parsed by a Perl script to >generate a table of H-bonds for the SQL database. > So a particular pdb file generates a particular set of H-bonds in a table that is stored in the database?
Then why not just send the table to the web page along with the rest of the information you send? It shouldn't be hard for you to do a little server-side or client-side manipulating of that to create the Jmol script necessary to do the connections. It would look like: connect (atomno=3430) (atomno=2234) hbond; connect (atomno=3431) (atomno=2567) hbond; connect (atomno=3432) (atomno=2345) hbond; connect (atomno=3433) (atomno=2134) hbond; connect (atomno=3434) (atomno=2245) hbond; Or, for that matter, you could append the hydrogen bonding information to the PDB file in the proper PDB format for hydrogen bond CONECT records, and then I think Jmol would (or should) just handle that directly. Should be quite feasible. In fact, if you wanted to, you could just have Jmol not do the hbonding and take all the hbonds from the table sent from the server, if it were important to display the HBPlus determinations precisely. Bob > The motif >definitions are cast as SQL queries to populate the database with >them. (They could be done on the fly, but it would be too slow.) When >a user makes a query for a particular motif in a particular, the >residues are retrieved (as residue numbers), and this info is used to >generate a pop-up page with the appropriate PDB file displayed in >Jmol and the buttons scripted to to highlight or restrict the display >to the motif. > >Jmol calculates mainchain-mainchain H-bonds on the fly, so if the >display is restricted to the motif residues, they show fine, so if >Jmol could also calculate the other H-bonds that would be ideal. One >alternative would be to query the db for the HBonds associated with >the motif and retrieve these, but this would add complexity and might >slow things down. The other might appear to be to compute the >sidechain-mainchain H-bond for a particular definition. This might be >Ok for eg the Asx Turn in which an Asx O at residue 1 is H-bonded to >a mainchain NH at residue 3, but for other motifs the mainchain >residue can be one of two positions. > >However Angel suggested: > > > >>say, >>connect 3.0 (AS?.N??) (SER.O??) hbond >>to hbond creates between nitrogen atoms in Asx and oxygen atoms in >>serine if they are less than 3 angstroms apart >> >> > >Which might be more promising, but would require generate pages with >different jmol scripts depending on the motif. > >Sigh. There's probably a good reason why none of the protein graphics >programs calculate the mainchain-sidechain H-bonds. > >David > > > > ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Jmol-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jmol-users
