This example uses example Cimetidine data from the EXPO software
|Before doing anything with the data (peak profiling, powder indexing, etc), it might be a good idea to see if it may already have been solved; and/or what related structures may exist and look like. Thus we will use the expected molecular connectivity to find out what related structures have already been solved.
Go to the CORINA page
and scroll down until you see the data entry for the SMILES string.
Select the Create Molecule button to bring up the JME Molecular Editor and
enter the fragment in 2D format. You may like to play with just entering a part of
the structure - or the entire structure - and see what the optimisation algoirthm
does. Here we have entered the entire molecule.
Now Submit the fragment to be minimized into a 3D conformation; and
viewed with the JAVA applet (smile string it uses being CNC(=NC#N)NCCSCc1nc[NH]c1C).
Save the now 3D fragment into PDB file ready to import into
various crystallographic software. In this case, directly import into
Platon to use it's ability to generate Quest scripts.
(Note: Ortep-3 by Louis Farrugia can import a PDB file and save it as a Shelx file in a ready to edit format.)
Now it is time to check out the fragment in the Cambridge Database, using Platon as
a user-friendly interface.
If you are already running Platon on a UNIX computer where the CSD and Quest already been installed, this is extremely trivial.
In this example we will deal with the case where the user is on a Windows PC, and the CSD database is on a remote UNIX machine where Platon has been compiled. (it is possible to generate the Quest query on Platon for Windows and ftp it over to run manually in quest - using a command line similar to quest -j structure_name < structure_name.que)
NOTE: With quest, make sure it is not automatically going into Xterm/Menu mode or the following will not work. By default, Quest should not go into Xterm/Menu mode but the administrator may have set it up like this to save on a few keystrokes.
If you have not done already, information on settup up for Secure FTP and Secure X sessions from a Windows machine, refer:
If a UK based academic or student, you can obtain free and easy access to the Cambridge and other structure databases via the EPSRC funded CDS - Chemical Database Service (free registration is done On-line)
FTP the PDB file to the remote machine. In this case the Chemical Database Service server (freely available after registration for any UK based academic or student) at cds.dl.ac.uk. (you will need your CDS username and password handy to do this)
Run teraterm and login to the UNIX computer containing the Cambridge database
(in this case cds.dl.ac.uk).
Run your Windows X-server (in this case the MI/XServer for Windows)
In the teraterm shell, go to the directory where the Shelx INS file with the starting information is, then type platon cime.pdb. This will spawn the following window in your windows X-server.
To redraw a screen if you swapped into a Window program, just use CONTROL L.
Following is how the structure looks in Platon using Pluton, then ADP (ORTEP) mode. (Just use the options under Graphics to do this).
Under the REPORT option, select CSD-QUEST and Platon will
generate the Quest file then try and spawn Quest if this is available locally.
If you swap into the Teraterm screen, you can see Quest running finding hits!
Platon will now display the structures that were found graphically.
As this shows that Cimetidine has already been solved, doing this first could save you a lot of bother.
In the present April 10th version of Platon, use the bottom right PREV and NEXT menu option to go to the next structure if it found more than one. This can include structures with interesting co-ordination with metals.
If platon tells you it got hits (on the bottom left of the screen) but does not give display a structure, but gives the following output at the bottom right of the screen. It could be there is a nuance with Platon and the Windows X-server in being able to spawn extra screens. Look manually in the Quest files, and these will tell you want structures it found.
Even if things go well, you can of course view manually the jnl journal file.
You can do the following manually in Quest by copying the *.QUE Quest file
generated by Platon to a machine with Quest on it and run quest and import the file.
quest -j structure_name < structure_name.que
QUEST Query file