MRV & SDF Format effect in the standardizer

Hi,

We are making a standardization of our database.  We tested to standardize these structures in SDF file and got a suitable config file. However, the different result was got by standardizing the same structure file in MRV . Could you help to check the config file?

In attached file of 'error 11': the right structure was standardized by the left one. the resulted structure was not what we need and seemed to be out of order

in attached file of 'error 12': the wrong right structure was got by standardizing the left one, but we never added the corresponding set in the config file 

How to avoid these unwanted result in MRV file?

Thank a lot

We will answer your question in a few days. Thank you for patience.

I am not sure, that I understand clearly which structures were your input. I assume, that the molecules on the left of your sample files. In this case, I cannot reproduce the issues using your config file. None of the molecules are affected by the standardization.

Can you please explain which structures were the input and what output did you expect?

Hi,Gyuri

Our structure database was bulit by jchem-base in Mysql. The version was 5.1.4. We added  the config file to our database so that the stucture can be inserted by standardized structure.The following script was set for standardization.

"String molfiles=request.getParameter("molfile");
String molfile=null;
Standardizer sta = new Standardizer(new File("config.xml"));
Molecule mols=MolImporter.importMol(molfiles);
Molecule molfilea=sta.standardize(mols);
molfile=(String)molfilea.toObject("mrv");"

The left structure in the attached file (error11&error12.mrv) was  input structure and the right one was resulted standardized stucture in the database. We wish the left stucture can be kept original shape but changed to the right structure.

Could you tell us what wrong with our script and config file? or what need to do ?

 

Thanks fo rthe explanation. Since I could not reproduce the issue with the current Standardizer application (GUI), we will try to reproduce it with the given version using the API next week. However, in the meantime, I would suggest you to upgrade (if possible).

Gyuri wrote:

Thanks fo rthe explanation. Since I could not reproduce the issue with the current Standardizer application (GUI), we will try to reproduce it with the given version using the API next week. However, in the meantime, I would suggest you to upgrade (if possible).

Many thanks for suggestion. We will take into account the jchem upgrading. 

We downloaded the latest jchem for JAVA(5.3.5) and used the Standardizer to standardize the above structure(input and output structures were in mrv format). We were pleased to find the problem of error12 was solved. But the problem of error11 can not be solved.  All the structure included "CF3, Boc, Cbz, ..."groups were out of shape. we tryed to delete the "contract groups" and get a better result. Could you help us to test if there is some bug with "contract group"? If we need to modify our config file, please let us know

Thanks

Hi,

We have identified the bug; the Molecule.contractSgroups() method messes up the coordinates somehow. We will notify you in this forum topic when the bug will be fixed.

Zsolt