Problem with the clean option in Standardizer

Hello,


I have a question about the clean option in the Standardizer 3.1.1. I joined in this post a file "molecules_cleanpb.sdf" that contain some big molecules I wanted to clean. I used the command line below:


standardize -c "clean:full" molecules_cleanpb.sdf -o molecules_cleanpb_std.sdf


that gave me lots of Warnings of this type:


"Warning: Cis-trans information changed at atoms 89 48 in molecule ([#8]C1CCCCC(=O)NC2=CC=CC=C2C3=C4NC(C=C4


)=C5C6=NC(C=C6)=C(C7=CC=C(N7)C(=C8C=CC3=N8)C9=CC=CC=C9NC(=O)NC%10=CC=CC=C%10C%11=C%12C=CC(=N%12)C%13=C%14NC(C=C%14)=C(C%15=NC(C=C%15)=C(C%16=CC=C%11N%16)C%17=C(NC(=O)CCCCC([#8])CCCC(=O)NC%18=CC=CC=C%13%18)C=CC=C%17)C%19=CC=CC=C%19NC(=O)NC%20=CC=CC=C5%20)C%21=CC=CC=C%21NC(=O)CCC1)


"


and I obtained stranges structures... I also tried the configuration "cleanstereo..clean:full" that doesn't give me Warnings and structures was a little different, but in both cases I have problem after, when I tempt to calculate the Major Microspecies at pH 7.4 with the Calculator 4.0.1 and the command line below:


cxcalc -N hi -v -o molecules_cleanpb_std_mjms.sdf majorms -H 7.4 -f sdf molecules_cleanpb_std.sdf


The message error is the following one:


Calculation warnings:





1: majorms:


2: majorms:


3: majorms:


4: majorms:


5: majorms:


6: majorms:


7: majorms:


8: majorms:


9: majorms:


and the file obtained is empty. I'm not sure that the reason is the clean option because I have the same problem with another molecule given in the post "Plugin Major Microspecies: somes molecules eliminates" in the section "Structure based prediction" of the forum but perhap's there is a connection.


Could you help me to undrestand this?


Best regards,


Samia ACI

Regarding the microspecies problem, the reason is the same as in your other post:


http://www.chemaxon.com/forum/viewtopic.php?t=973


Can you see the warning text? For me this warning is written to the console:


majorms: The number of ionizable atoms (9) exceeds the specified limit (8).


For the reason, see my answer in


http://www.chemaxon.com/forum/viewtopic.php?t=973





Regarding the clean problem, probably the reason is that the cis-trans data could not be maintained during the clean process - this is most likely a problem in our clean algorithm rather that in your input file - but we will give a more precise answer to this.

Thanks a lot for your answer. Strangely the entire message didn't appears on my console, I only saw the line


1: majorms:


so I didn't understand the reason and I ask you the question.


I am also surprised because, when I visualise the molecule in MarvinView (4.0.1) and then, use the button Tools>MajorMicrospecies, I have no message error and a ionized molecules appears in a pop-up window while if I use the command line, the molecule is never generated. How is it made?


Thanks, regards


Samia ACI

Sorry for the late answer, I have noticed your comment here just now.





I could reproduce the error message display problem with the previous MarvinBeans - this has been fixed since then. Also the major microspecies are now generated (although a bit slowly) for any number of ionizable atoms - if there are more than 8 ionizable atoms then we use a different (and slower) method for generating the microspecies.





These corrections are included in the latest release Marvin 4.0.2, you can download it from:


http://www.chemaxon.com/marvin





Remarks:





1. Although the Marvin GUI did not show the error message in the previous release Marvin 4.0.1, the generated molecule was not the major microspecies - the reason is the bug in the error message: since there was no error message, the result moelcule was queried and shown and the error was not realized.





2. The major microspecies are shown in standardized form for technical reasons, which means sulphynil and nitro group standardization and aromatization in most cases. However, in SDF format it is not correct to output aromatized target (non-query) molecules. Therefore it is better to dearomatize the molecules before output with output option -f sdf:-a - also, the option -N hi is not needed any-more because that is the default for plugins with molecule (non-text) output:

Code:

cxcalc -v -o molecules_cleanpb_std_mjms.sdf majorms -H 7.4 -f sdf:-a molecules_cleanpb_std.sdf

Thanks a lot for your answer and to have informed me about the new release of MarvinBeans. I use this version 4.0.2 of the calculator plugin now and my molecules which have more 8 ionizables atoms are no longer a problem but unfortunatly it appears an other problem when I use new molecules, the problem is reported below with the SMILE code of one of my molecules (this time, with the new release, I can report the entire warning message, thanks!):


cxcalc -v majorms -H 7.4 -f smiles "CCOC(=O)C1(CC[N+](CC1)(C)[O-])C2=CC=CC=C2"





1: chemaxon.marvin.plugin.PluginException: Inconsistent molecular structure.





Calculation warnings:





1: majorms: Inconsistent molecular structure.


Could you another time bring me your assistance on this error?


Thanks, best regards


Samia

This was a bug in the code and not in the molecular structure. The bug has been fixed in the next bugfix release Marvin 4.0.3. This will be available by tomorrow at


http://www.chemaxon.com/marvin





Thanks for the report.