How to use JChem Base for structure searching?

How can I use JChem Base for structure searching?

Basically there 3 options to use JChem Base for structure searching:





1) Using the core (low-level) functions in the Java API (http://www.jchem.com/doc/api/index.html)





Some important classes:


Molecule: http://www.jchem.com/doc/api/chemaxon/struc/Molecule.html


MolSearch (for exact or substructure searching: see the example in the javadoc): http://www.jchem.com/doc/api/chemaxon/sss/search/MolSearch.html


GenerateMD (generating fingerprints for similarity and fast substructure searching): http://www.jchem.com/doc/api/chemaxon/descriptors/GenerateMD.html


If you want to use fingerprints, then you may need more pointers. Let us know that.





For efficient structure searching, create a cache of the structures. Use the SMILES format (http://www.chemaxon.com/marvin/doc/dev/formats.html) and possibly fingerprints in the cache.





2) Using the database functions in the Java API





In this case you don't need to bother with fingerprints and caching. Everything is handled in these high-level classes. However, you will need a JDBC/ODBC connection to a databases (like Oracle, DB II, MySQL, PostgreSQL, ACCESS, ...).


JChemSearch (for all types of structure searching: see the example in the javadoc): http://www.jchem.com/doc/api/chemaxon/jchem/db/JChemSearch.html


UpdateHandler (database manipulations): http://www.jchem.com/doc/api/chemaxon/jchem/db/UpdateHandler.html


Importer: http://www.jchem.com/doc/api/chemaxon/jchem/db/Importer.html


Exporter: http://www.jchem.com/doc/api/chemaxon/jchem/db/Exporter.html





Example application


Play: http://www.jchem.com/examples/jsp1_x/index.jsp


Setup: http://www.jchem.com/examples/jsp1_x/setup.html





Example codes: http://www.jchem.com/doc/guide





3) Oracle cartridge (Everything from SQL):


http://www.jchem.com/doc/guide/cartridge/index.html





--





Other aspects:





Marvin:


You may also need Marvin for structure drawing and visualization: http://www.chemaxon.com/marvin/


It is included in JChem Base as


- JavaBeans


- Java Applets


- Applications





Search speed: http://www.jchem.com/FAQ.html#benchmark





Size in memory: 3 million structures (SMILES and fingerprints) <<300MB, see http://www.jchem.com/FAQ.html#outofmemory


You can also store prepared (standardized/aromatized) Molecule objects to improve speed, but they are much bigger than SMILES)





Standardizer:


You may need to canonize the structures for proper searching/visualization: http://www.jchem.com/doc/user/Standardizer.html