Synthesizer in Database Mode

Hi,





I was trying to follow the synthesizer example using linear in database mode.


When I run the first few commands from the windows prompt, I do not get any errors. When I look directly at the database, I see new tables created and data stored. When I try to view the structures in JChemManager with Synthesizer I do not see any structures and it reports zero. After I run the synthesize command I get an Invalid Fetch Size error.





I am using an academic license. Below is the sequence of commands I entered on a Windows XP machine.





C:\Program Files\ChemAxon\JChem\examples\synthesizer\linear>synthesize c -n syn


-c Synthesizer.xml


C:\Program Files\ChemAxon\JChem\examples\synthesizer\linear>synthesize i -n syn


-s scaffold scaffold.smiles


C:\Program Files\ChemAxon\JChem\examples\synthesizer\linear>synthesize i -n syn


-s alkyne alkynes.smiles


C:\Program Files\ChemAxon\JChem\examples\synthesizer\linear>synthesize i -n syn


-s amine amines.smiles


C:\Program Files\ChemAxon\JChem\examples\synthesizer\linear>synthesize i -n syn


-s carboxylic-acid carboxylic-acids.smiles


C:\Program Files\ChemAxon\JChem\examples\synthesizer\linear>synthesize r -n syn


java.sql.SQLException: Invalid Fetch Size


at sun.jdbc.odbc.JdbcOdbcStatement.setFetchSize(Unknown Source)


at sun.jdbc.odbc.JdbcOdbcResultSet.setFetchSize(Unknown Source)


at chemaxon.reaction.SynthesisSetReader.fetchIDsIfNeeded(SynthesisSetRea


der.java:441)


at chemaxon.reaction.SynthesisSetReader.nextWithPrefetch(SynthesisSetRea


der.java:456)


at chemaxon.reaction.SynthesisSetReader.next(SynthesisSetReader.java:116


)


at chemaxon.reaction.synthesis.DbSetContent$DbMolIterator.next(DbSetCont


ent.java:107)


at chemaxon.reaction.synthesis.MultiMolSetCounter.next(MultiMolSetCounte


r.java:107)


at chemaxon.reaction.synthesis.LinearAlgorithm.step(LinearAlgorithm.java


:204)


at chemaxon.reaction.Synthesizer.step(Synthesizer.java:849)


at chemaxon.reaction.Synthesizer.run(Synthesizer.java:823)


at chemaxon.reaction.Synthesizer.dbmain(Synthesizer.java:1245)


at chemaxon.reaction.Synthesizer.main(Synthesizer.java:1424)


chemaxon.reaction.ReactionException: Invalid Fetch Size


chemaxon.reaction.ReactionException: Invalid Fetch Size


at chemaxon.reaction.synthesis.DbSetContent$DbMolIterator.next(DbSetCont


ent.java:110)


at chemaxon.reaction.synthesis.MultiMolSetCounter.next(MultiMolSetCounte


r.java:107)


at chemaxon.reaction.synthesis.LinearAlgorithm.step(LinearAlgorithm.java


:204)


at chemaxon.reaction.Synthesizer.step(Synthesizer.java:849)


at chemaxon.reaction.Synthesizer.run(Synthesizer.java:823)


at chemaxon.reaction.Synthesizer.dbmain(Synthesizer.java:1245)


at chemaxon.reaction.Synthesizer.main(Synthesizer.java:1424)


Caused by:


chemaxon.reaction.SynthesisSetReaderException: Invalid Fetch Size


at chemaxon.reaction.SynthesisSetReader.next(SynthesisSetReader.java:146


)


at chemaxon.reaction.synthesis.DbSetContent$DbMolIterator.next(DbSetCont


ent.java:107)


at chemaxon.reaction.synthesis.MultiMolSetCounter.next(MultiMolSetCounte


r.java:107)


at chemaxon.reaction.synthesis.LinearAlgorithm.step(LinearAlgorithm.java


:204)


at chemaxon.reaction.Synthesizer.step(Synthesizer.java:849)


at chemaxon.reaction.Synthesizer.run(Synthesizer.java:823)


at chemaxon.reaction.Synthesizer.dbmain(Synthesizer.java:1245)


at chemaxon.reaction.Synthesizer.main(Synthesizer.java:1424)

My collegue will answer tomorrow.





I checked your profile, and it seems, that you are interested in biotransformations (though the website you specified is something very unrelated ;) ).


If you would like to predict biotransformations, and not generating combichem libraries, we might have some interesting alternatives for you.





FYI, we plan to substitute the current synthesizer tool with some advanced Instant JChem modules later in this year.

Yes, the site was unrelated as I had to place some early experiments under a test site. We have since moved the work to www.retro-biosynthesis.com. Thanks for mentioning it so I updated that element of the profile.





I would love to hear more about different approaches. We are trying to build a publicly accessible tool for researchers interested in bio-transformation and in particular a new approach to de novo designs.





We have posted the first element of the approach. We are now trying to develop some rules for pathway design efforts that would help optimize the experimental approaches.





Prof. Prather at MIT is leading the effort.

Hi,





What is the database management system you use?


What Java version do you use?





Zsolt

MySQL on a unix box via DSN from Windows XP.

Hi,





ODBC connection is currently only supported / tested for Microsoft Access.


(We will update the documentation about this)


I suggest you to connect directly via JDBC connection.


You will need the JDBC driver in your CLASSPATH environment variable for this.





I hope this help, let us know if you have further questions.





Best regards,





Szilard

Thanks, that did the trick.





I changed the driver from the


sun.jdbc.odbc.JdbcOdbcDriver


to


com.mysql.jdbc.Driver


in JChemManager.





I copied the driver's jar to the lib/ext director of my Java version and set my system CLASSPATH to that location. JChemManager would then log in correctly and I could view the structures that I had uploaded. However when I ran


synthesize r -n biodegradation


I got a class not found error for the driver.





I then copied the driver to


C:\Program Files\ChemAxon\JChem\lib





After doing that


synthesize r -n biodegradation


ran without any errors.





I was also able to see everything using the


synthesis browser.





Thanks much for your help.





Based on the earlier posting it appears that you are dropping synthesizer, is this correct?











Darcy

Hi Darcy,

Quote:

However when I ran synthesize r -n biodegradation I got a class not found error for the driver.

I suspect it was because you have tried to run it in a previously opened console.


After changing system variables new consoles should be opened.





The location of the JDBC driver is arbitrary as long as it is in the CLASSPATH, there is no need to copy it to any special location.





Szilard

Quote:

Based on the earlier posting it appears that you are dropping synthesizer, is this correct?

We are about to substitute the different functionalities of the generic synthesizer tool with new, specific classes. For example, we are developing Metabolizer, which will predict human xenobiotic metabolites and metabolic stability. Furthermore we will develop a combichem library generation modul for Instant JChem in the near future.





These upcoming new software tools are based on the Reactor technology similarly to Synthesizer, but will offer advanced specific functionalities and easy to use user interface.

Hi,


Sounds interesting. Just to be precise, are you dropping synthesizer entirely?


If I do some development with it now does it mean that it will not survive future upgrades?





Darcy

We will not remove Synthesizer from the class library in the near future, however, we do not plan to enhance it.





I suppose, that you might consider developing your library generation application using Reactor directly, you can certainly enjoy our technical support. If you are working on metabolic biotransformations, the upcoming Metabolizer API can also be interesting.





If you need more info or have questions, we might arrange a webex.