Example in the query feature document

http://www.chemaxon.com/jchem/doc/user/query_features.html





In Table 3, the 4th query does not seem to match the 2nd target.

It should be OK.





Please note that QH (hetero or hydrogen) matches the implicit H.





Regards,


Szabolcs

Hi,





It matches because of the presence of the implicit H (see attached picture).





Best regards,


Tamás

Thank you. Just out of curiosity, I tried jc_compare


The following statement returns 0, no match.


I also try to represent query in MOL format, same result. I must have a mistake somewhere.





select jc_compare('CC(Cl)C(O)C1CCCC(C1)C(C)CBr', '[#6]C([!#6])CBr', 't:s') FROM dual;

Hi,





In SMARTS format these should not match, as only isotopic and charged Hydrogens match not list according to Daylight conventions.


Please see:





http://www.chemaxon.com/jchem/doc/user/query_appendix.html#daylMDLDiff





However in mol format there should be a match according to the documentation.





Could you attach the files in mol/SDF format and let us know you JChem version please ?





Best regards,





Szilard

<pre>





Marvin 11110808542D





5 4 0 0 0 0 999 V2000


-1.5616 0.3241 0.0000 QH 0 0 0 0 0 0 0 0 0 0 0 0


-0.8471 0.7366 0.0000 C 0 0 0 0 0 0 0 0 0 0 0 0


-0.1327 0.3241 0.0000 C 0 0 0 0 0 0 0 0 0 0 0 0


0.5818 0.7366 0.0000 Br 0 0 0 0 0 0 0 0 0 0 0 0


-0.8471 1.5616 0.0000 C 0 0 0 0 0 0 0 0 0 0 0 0


1 2 1 0 0 0 0


2 3 1 0 0 0 0


3 4 1 0 0 0 0


2 5 1 0 0 0 0


M END


</pre>

Hi,





You did not specify the JChem version, so I was trying with one of the latest.


(Please also attach mol files as files, it is much easier to manage them)





The not list query in mol file format works as expected.





The mol query you have pasted on the other hand contains an QH query atom.


This seems like a bug, my colleagues will investigate this and get back to you.


Until then you may use the not list in mol format for the same purpose.





Best regards,





Szilard

I wanted to post MOL as a file, but believe it or not, I had a very weird situtation. I use Marvin Sketch to save the MOL file, the MOL file is visible within Marvin Sketch Open dialog, but not visible to the Windows Explore (when saved under Desktop or My Document). So I was not able to attach the MOL file. Very strange indeed. My colleague had exactly the same issue last week. We are using Vista Ultimate X64. I was doing some quick Google search, did not find useful link. Anyway, I will try more when I get time.





I was using Marvin in jchem_5_0_02_1. But I just tried the example on the ChemAxon web site, when I try to get the query MOL format, it uses QH as element name as well.

The first issue seems to be some strange Windows "feature".





Regarding the conversion, I'm not sure I understand you correctly.


If I take the query SMARTS "[#6]C([!#6])CBr" and paste it into a Marvin Sketch (e.g. one of the applets on our website) I get the "not list" format on either file export or in the Edit -> Source window when I select mol format in "View".


It applies for both V2000 and V3000 mol formats.


Maybe You have converted an other SMARTS string (they all look alike to me :) ) or you have used some other method ?





In any case the search behavior of the QH must be investigated regardless.





Szilard

I did not write the smarts, I actually draw the query using "QH" atom, then obtain the MOL file and use the MOL in the jc_compare search. If you draw the query (Table 3) and try to get MOL, you will see QH in the atom list.

Hi Yingyao,





You are right, our SMARTS export is buggy, [!#6] is not the correct representation for QH. We will change that to [#1,!#6].





As a workaround, you could use cxsmarts instead, that preserves the QH query atom:

Code:

[#6]C([!#6])CBr |$;;QH_p;;$|

Best regards,


Szabolcs

Just want to make sure, it sounds the MOL export for QH atom is buggy as well.

Hi,

Quote:

Just want to make sure, it sounds the MOL export for QH atom is buggy as well.

Yes it seems that is the problem that's preventing you from getting the desired search result.





We are investigating this issue and will get back to you soon.





Until then as a workaround you may use the MRV format for searching with QH query atoms.





Best regards,





Szilard

I did a bit more Google search, first I need to convince you that I am not out of my mind. Others have similar problem with the ghost files.





http://www.vistax64.com/vista-general/41666-java-6-creates-directories-files-explorer-cant-see.html





The page below does not really explain what I saw (because I saved the file in Desktop, and my Document, not into Program Files folder). However, it seems offer a good theory - the MOL file Marvin Sketch wrote was actually saved in another location. I have to uninstall Java 6 Upgrade 10, I cannot reproduce the problem at this point. (I guess it has something to do with Vista x64 + Java 6 Upgrade 10).





http://stackoverflow.com/questions/280600/java-filecanwrite-on-vista-and-super-hidden-files





I post the info here just in case someone else experience the same.

Although I cannot reproduce it, I believe the answer is AUC.





http://www.eggheadcafe.com/software/aspnet/32468146/cant-see-swf-files-save.aspx





Why Java can create a file that people in the Administrator group cannot see (but should be visible by Administrator account)? I don't know.

Quote:

Just want to make sure, it sounds the MOL export for QH atom is buggy as well.

Actually the bug was in the MOL import, not the export.


We have fixed it, the fix will be available in the next minor release (JChem 5.1.4)





Thank you for the bug report.





Best regards,





Szilard

Do you have a rough estimation on when 5.1.4 will be released? Thanks

Hi,





It is planned to be out next week.





Best regards,


Szabolcs