Cannot index simple molecule

Hello!

    I receive an error trying to index a H₂ molecule with the JChem Cartridge v6.1.4. The file with the molecule is in the attachment, the stacktrace of the error is:

[RMI TCP Connection(36)-127.0.0.1/74]: 2014-01-22 13:39:54.786 chemaxon.jchem.cartridge.rmi.impl.MolSearchImpl targetStringToMol
SEVERE: Offending structure: '[B@10304d3'
chemaxon.formats.MolFormatException: Cannot read molecule file.
    at chemaxon.formats.MolInputStream.init(MolInputStream.java:146)
    at chemaxon.formats.MolInputStream.<init>(MolInputStream.java:115)
    at chemaxon.formats.MolInputStream.<init>(MolInputStream.java:57)
    at chemaxon.util.MolHandler.importMol(MolHandler.java:672)
    at chemaxon.util.MolHandler.setMolecule(MolHandler.java:179)
    at chemaxon.util.MolHandler.<init>(MolHandler.java:111)
    at chemaxon.jchem.cartridge.rmi.impl.MolSearchImpl.targetStringToMol(MolSearchImpl.java:277)
    at chemaxon.jchem.cartridge.rmi.impl.MolSearchImpl.search(MolSearchImpl.java:92)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:601)
    at sun.rmi.server.UnicastServerRef.dispatch(UnicastServerRef.java:322)
    at sun.rmi.transport.Transport$1.run(Transport.java:177)
    at sun.rmi.transport.Transport$1.run(Transport.java:174)
    at java.security.AccessController.doPrivileged(Native Method)
    at sun.rmi.transport.Transport.serviceCall(Transport.java:173)
    at sun.rmi.transport.tcp.TCPTransport.handleMessages(TCPTransport.java:553)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run0(TCPTransport.java:808)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run(TCPTransport.java:667)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
    at java.lang.Thread.run(Thread.java:722)
Caused by: chemaxon.formats.MolFormatException: Cannot recognize format (?)
Unrecognized file contents:
 
    at chemaxon.formats.recognizer.RecognitionSubsystem.getFormat(RecognitionSubsystem.java:219)
    at chemaxon.formats.MolInputStream.initTextFormat(MolInputStream.java:315)
    at chemaxon.formats.MolInputStream.init(MolInputStream.java:135)
    ... 22 more
[RMI TCP Connection(36)-127.0.0.1/74]: 2014-01-22 13:39:54.786 chemaxon.jchem.cartridge.rmi.impl.RmiExceptionHandler handleError
SEVERE: Invalid target: Cannot read molecule file.
chemaxon.jchem.cartridge.sharedorajcsrv.rmi.StructureFormatException: Invalid target: Cannot read molecule file.
    at chemaxon.jchem.cartridge.rmi.impl.MolSearchImpl.search(MolSearchImpl.java:113)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:601)
    at sun.rmi.server.UnicastServerRef.dispatch(UnicastServerRef.java:322)
    at sun.rmi.transport.Transport$1.run(Transport.java:177)
    at sun.rmi.transport.Transport$1.run(Transport.java:174)
    at java.security.AccessController.doPrivileged(Native Method)
    at sun.rmi.transport.Transport.serviceCall(Transport.java:173)
    at sun.rmi.transport.tcp.TCPTransport.handleMessages(TCPTransport.java:553)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run0(TCPTransport.java:808)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run(TCPTransport.java:667)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
    at java.lang.Thread.run(Thread.java:722)

Thank you.

Hello Leonid,

This structure below is an invalid structure. What does it want to represent?

Offending structure: '[B@10304d3'

 

 Krisztina

Hello Krisztina,

    I guess the output we see is a call to toString() for some byte array happening somewhere inside JChem so I don't know what it is. What I try to index is the structure attached to the original post.

Thank you.

Hello Leonid,

Could you send us all the parameters of the index creation including the applied standardization actions?

Thank you

Krisztina

Hello Krisztina,

I'll be a follow-up to Leonid on this issue. This is how we create our structure table (CRP_STRUCTURE):

Execute JChem_Table_Pkg.Create_JCTable('CRP_STRUCTURE', 
   'JCHEMPROPERTIES', 16, 2, 6, ', MOLFILE CLob', Null, 1, 'tableType:molecules');

and its index:

Create Index IDX_CRP_STRUCTURE On CRP_STRUCTURE(CD_STRUCTURE) IndexType 
   Is JCHEM..JC_IdxType;

We issue the following command to insert into this table (our JChem user is ELN_JCHEM):

JChem_Table_Pkg.JC_Insert(


                      'Select CONVERSION_BLOB.CNV_OBJECT' ||


                      ' From  CONVERSION_BLOB' ||


                      ' Where CONVERSION_BLOB.CNV_ID = ' || in_UserSession


                    , 'ELN_JCHEM..CRP_STRUCTURE'


                    , Null


                    , 'true', 'false'


                    , 'flags:c haltOnBadFormat:n'

                );

Does this help in any way?

 

Thank you,

Vitaly

Hi Vitaly,

We've tracked down this issue and have now the following assumptions:

• In your original standard structure table, there is at least one non-valid chemical structure.


• when this table was indexed with haltOnError:n (or nf)  or haltOnBedFormat:n parameter, the non-valid structure was also indexed, and in the index table, its cd_hash field was filled with '0' value.


• the cd_hash value of H2 molecule is also '0'


• when you try to insert H2 molecule using the 'duplicate filtering' option, in the background a duplicate search runs through the index table; during this search the content of the cd_hash field is also compared


• in case of the H2 molecule and the stored non-valid structure a conflict is generated because hashmarks are the same and the molecules are not

At present, our software is not prepared for such situations, we only can recommend not to use duplicate filtering and haltOnError:n (nf) or haltOnBedformat:n  parameters at same time.

Best regards,

Krisztina

Hello Krisztina,

Is it possible that adding a no-structure molecule can cause this problem? Currently we are reproducing it starting with a clean database, adding a no-structure and after that adding H2.

 

Thank you,

Vitaly

Hi Vitaly,

Empty structure (no-structure)  has also '0' as hash value, but - at least at JChem level - this case is handled and H2 can be imported into a JChem table containing empty structures and bearing 'filter out duplicate structures' table setting. We'll check now how does JChem Cartridge handle this case.

Krisztina

Hi Vitaly,

Unfortunately, at the moment it is really not possible to insert H2 molecule into a JChem table using

jchem_table_pkg.jc_insert 

function of JChem Cartridge. It seems to be bug that will be registered, and later planned when to be fixed.

Until then, we suggest to use directly JChem for importing H2 molecule into JChem tables (e.g. JChem Manager GUI).

Best regards,

Krisztina

 

Hi Krisztina,

 

Thank you very much for investigating this issue, we'll be keeping an eye on JChem updates.

 

Vitaly

Hi Vitaly,

Unfortunately, we have not fixed this issue yet and it is likely that we will not have time to fix it. (On the other hand it seems that not much people registering H2 molecule in their database.)

So, if you have resolved this issue with a workaround I would remove this issue from our bug tracking system.