InstantJChem batch import

This is a question about batch importing .SDF files. Can this be done with instantJchem (development version 200610171010) or should it be done with the Jchemmanager. If it should be done with the Jchemmanager, why can't I connect to the mysql instantjchem database that I created using the same jdbc driver/url setup that I am using with instantjchem to access that database with jchemmanager?





Settings for connecting with Jchemmanager/instantJchem:





driver: com.mysql.jdbc.Driver


URL: jdbc:mysql://localhost/MeOHconifer


property tabe: JChemProperties


login: me


pswrd: **





I have now also been playing with the molconverter and RDF import in instantJchem however I get an error when I click finish after trying to create a template:





java.lang.NullPointerException


at com.im.workbench.ijc.io.rdfimport.TemplateBuilder.safeDBName(TemplateBuilder.java:136)


at com.im.workbench.ijc.io.rdfimport.TemplateBuilder.createTemplate(TemplateBuilder.java:230)


at com.im.workbench.ijc.io.rdfimport.wizard.RdfimportWizardIterator$TemplateBuilderRunner.buildTemplate(RdfimportWizardIterator.java:291)


at com.im.workbench.ijc.io.rdfimport.wizard.RdfimportWizardIterator.createTemplate(RdfimportWizardIterator.java:193)


at com.im.workbench.ijc.io.rdfimport.wizard.RdfimportWizardIterator.instantiate(RdfimportWizardIterator.java:169)


at org.openide.WizardDescriptor.callInstantiate(WizardDescriptor.java:1203)


at org.openide.WizardDescriptor.access$1600(WizardDescriptor.java:76)


at org.openide.WizardDescriptor$Listener$2$1.run(WizardDescriptor.java:1682)


at org.openide.util.RequestProcessor$Task.run(RequestProcessor.java:499)


[catch] at org.openide.util.RequestProcessor$Processor.run(RequestProcessor.java:932)








Thank-you,


Scott

Hi Scott,

Quote:

This is a question about batch importing .SDF files. Can this be done with instantJchem (development version 200610171010) or should it be done with the Jchemmanager.

You can import SD files dirrectly within Instant JChem. This is usually quite straight forward.


See the user documentation on the exact details.


http://www.chemaxon.com/instantjchem/ijc_2_0/docs/user/help/htmlFiles/moving_data/importing_data.html

Quote:

why can't I connect to the mysql instantjchem database that I created using the same jdbc driver/url setup that I am using with instantjchem to access that database with jchemmanager?

As a guess here I would say that this is the slightly strange behaviour in MySQL that treats table names in a case sensitive manner. Thus the property tables JCHEMPROPERTIES and JChemProperties are different tables in MySQL, but the same table in other databases. Instant JChem generally tries to work with all table names in upper case to minimise protential problems here.

Quote:

I have now also been playing with the molconverter and RDF import in instantJchem however I get an error when I click finish after trying to create a template: java.lang.NullPointerException at com.im.workbench.ijc.io.rdfimport.TemplateBuilder.safeDBName(TemplateBuilder.java:136)

RDF import is a pretty complex area. The RDF data format is poorly defined, resulting in it being used in all sorts of different ways. The current RDF import in Instant JChem should be viewed as a prototype to try to learn how to handle as many of these variants as posible. Its quite posible that your RDF file is one of these new variants that we haven't yet considered. If you can send us an example we will look at it (just a few entries from the file will be sufficient).





One simple thing though - did you specify the RDF file that you wanted to generate the template for in step 1?





Alternatively, there could be an error in the way you are trying to import the file. Try using the attached template to import the sample wombat data that is included in Instant JChem and can be downloaded from here:


http://sunsetmolecular.com/downloads/files/wb2004_histH1GPCRsRDF.zip


Put both files in the same directory and rename the attached file to wb2004_histh1gpcrs.properties (the forum does not allow this file extension to be used as an attachement). Then you should be able to import that RDF file. Alternatively you should be able to generate a template for that RDF file.


This should also give you an idea of how the templates work. As RDF import is a prototype the template definition is not yet documented. If you have any questions then ask in this forum.





Finally, from that error I can see that you are not using the latest updates. Load these from the Update Center (Tools menu) to ensure you have the latest version.





Tim

Hi Tim,





I've got a bunch of SDFs, like a 1000 or so, to import into an instantjchem under a single project. Ideally I'd like to be able to use your federated search function on all these SDFs after import, so then don't have to be in a single data tree, just in the same project.





How do you suggest I go about this?





Thanks and great program.





Andrew

There is currently no way to automate the import, so if you wanted separate tables for each file then you would have to manually import each file.





Your best solution is probably to combine the files into a single file and then import that. You can do this with the molconvert command line program. e.g.





molconvert -o all_structures.sdf sdf *.sdf





Tim