Menu
▾
▴
[Joelib-devel] MolConnZ, import and batch files under windows
|
From: Joerg K. W. <we...@in...> - 2004-02-19 09:17:35
|
Hi karthikeyan, > 1) thank you for your immediate response. I can wait till I can see the > atom symbols in 3D (i feel it is very much required for chemist to understand the > molecule in 3D view) > I will try the 2D technique suggested by you. I agree, but it the priority is still very low for me :-) > 2) Second query: (pl.) > I have problem in reading Molconnz (.s) file in Joelib. I get an error: > Molecule entry (#1) skipped: .. io.MoleculeIOexception: Line 14(1) > should contain 7 > descriptor not 11: 2.828427 2.00000 1.4142 1.000 0.0000 etc., > I am using the molconnz output directly This depends on your MolConnZ version ! The supported format is defined in joelib\src\joelib\data\plain\molconnz350.txt and the first lines should contain: id nvx nrings ncirc nelem fw aname ... so if you are using any other version you must supply your file definition, e.g. molconnz400.txt and define these descriptors in joelib\src\joelib\data\plain\knownResults.txt where you must define if your descriptors are double, integer, boolean, or whatever. If you do not define these things JOELib can still of course load the entries and select them, but always as string ! So if you plan to normalize your data or filters, or something else i recommend the definition in knownResults.txt. > or should I convert into some format before submitting to joelib? Is possible also, e.g. SDF which is my preffered format, but still not the best one, because you must still define the descriptors in knownResults.txt. The CML in JOELib is the most verbose format, because the descriptors obtains already a format (after defined in knownResults.txt )! Then they can be loaded without any definitions ! Furthermore the CML reader/writer is consistent in JOELib, but eventually not up-to-date with the Murray-Rust CML2 implementation in his Java library and OpenBabel, because he develops a huge amount of code i'm not able to follow so fast. And i'm not sure about their descriptor abilities, because OpenBabel has no descriptor storing facility. Please correct me anybody if this is not true. The conversion of XML files should be not to difficult, so ... > The final objective is: to read all molconnz descriptor and 'optionally' > write in a clean format (col/row) > > mol1 d1 d2 d3 d4.. > mol2 d1 d2 d3 d4.. SMILES/Flat file format is supported also in sh convertSkip.sh where the flat file format should be defined in a separate file format.txt with mol1-ID d1 d2 d3 d4 or in joelib.properties for the SMILES, but i think i've added a command line switch also!!! > 3) finally... regarding compiling using ant, the output is directed to > build directory > and how to run from main *.bat files if the output is in build > directory? > as a shortcut I copied all the *.bat files to build directory.. and it > is working ok As already discussed i do not like the bat-files and they are not really supported and up-to-date. If possible in any way, i recommend cygwin !!! A unix shell for windows, so you can use the shell scripts !!! Of course the bat files will work if all required libraries are added to the classpath (classes in build-directory and all lib/*jar files), but that's a boring work and changes often ... The shell scripts or ant will resolve the dependencies automatically ! Regards, Joerg > > regards > > > > -- > M. Karthikeyan, Ph.D., Scientist > _| _| _|_|_| _| > _|_| _| _| _| > _| _| _| _| _| > _| _|_| _| _| > _| _| o _|_|_|o _|_|_|_| > National Chemical Laboratory > Pune - 411 008, INDIA > Ph: +91-(0)20-5893 457 FAX: 5893 973 > http://www.ncl-india.org/ > > > -- Dipl. Chem. Joerg K. Wegner Center of Bioinformatics Tuebingen (ZBIT) Department of Computer Architecture Univ. Tuebingen, Sand 1, D-72076 Tuebingen, Germany Phone: (+49/0) 7071 29 78970 Fax: (+49/0) 7071 29 5091 E-Mail: mailto:we...@in... WWW: http://www-ra.informatik.uni-tuebingen.de -- Never mistake motion for action. E. Hemingway |
