wedge bond removal

Hi,


I used the Standardizer 3.2.9 funtions to get rid of *all* stereo information

Code:

<?xml version="1.0" encoding="UTF-8"?> <!-- Standardizer configuration file --> <!-- This configuration file is created with ChemAxon Config Builder --> <StandardizerConfiguration Version ="0.1">    <Actions>       <Removal ID="removal" Method="keepLargest" Measure="keepLargest"/>       <Dearomatize ID="dearomatize"/>       <ClearStereo ID="clearstereo"/>       <AbsoluteStereo ID="clearstereo" Act="Clear"/>    </Actions> </StandardizerConfiguration>

I set clear stereo to clear all and absolute stereo to remove all. Bust still I get stereo bonds in the results file. (Test data is the Biometa KEGG after cleaning with molconvert). Is there an additional Standardizer option to remove all stereo information?





Tobias

Using the API and the following code works see link below.





//according to Andras; http://www.chemaxon.com/forum/ftopic3093.html&highlight=


//clear all stereo


for(int i = 0; i<target.getBondCount(); i++){


MolBond b = target.getBond(i);


// remove parity info


b.setFlags(0,MolBond.STEREO_MASK);


// remove CIS TRANS info


int t = b.getType();


if (t == 2 || t == MolBond.SINGLE_OR_DOUBLE || t == MolBond.DOUBLE_OR_AROMATIC || t == MolBond.ANY){


b.setFlags(StereoConstants.CIS|StereoConstants.TRANS, StereoConstants.CTUMASK);


}


}





Tobias

The wiggly bonds above are not for tetrahedral stereo, but for unspecified double bond stereo. Without that, your double bonds would have stereo configurations specified by their geometry.

Gyuri wrote:

The wiggly bonds above are not for tetrahedral stereo, but for unspecified double bond stereo. Without that, your double bonds would have stereo configurations specified by their geometry.

Hi,


this is true, but isn't that about the level of specificity?


Because I can draw all of these molecules without any stereo information. And this is what I wanted from the Standardizer.





The problem is with our analytical techniques we can not


see the stereo information, but we can see the molecule.


So for us it is reasonable to point to certain structure with no stereo information at all.


This is the mess with all chemical databases, sometimes


you have mixtures or racemates and people say: yes thats


Glucose. Nobody is saying oh I had a teaspoon of


α-D-glucopyranose and β-D-glucopyranose and specify ring or chain form.





The problem with the current solution and the wiggly bonds is,


these molecules look uncommon and unfamiliar and ugly.


Even if (from a purist standpoint) the structure is ok.


I would like to have a function to remove the wiggly and wedge bonds at all.


Is there a way to do that with the Standardizer?





Kind regards


Tobias

Options for unspecified stereo double bonds:


- wiggly ligand (preferred by IUPAC)


- crossed double bond (older, but unpreferred by IUPAC)


- straight ligands (cannot place atoms straight in most cases)





Otherwise, how would you distinguish Z-butene from butene with unspecified stereo? If you remove the wiggly bonds from double bonds, they will all be E and Z isomers and you will loose the original stereo configuration data.





Anyway, I understand your point and will discuss it with my collegues. You probably need an option to turn off the wiggly double bonds and wedges as a display option! This solution might keep the stereo info of molecules and would not hurt your eyes. :)

We registered your request to turn off displaying the wedge bonds optionally in the View menu, and added it to our task list.


However, we are not sure about the general importance of this feature. So if anyone else would like to turn off the wedge bond display in the Marvin, please write a note here!

Gyuri wrote:

We registered your request to turn off displaying the wedge bonds optionally in the View menu, and added it to our task list. However, we are not sure about the general importance of this feature. So if anyone else would like to turn off the wedge bond display in the Marvin, please write a note here!

Hi Gyuri,


Sorry I was not very specific, the solution of no stereo information at all is not needed for beautification.





Rather I would have no cis/trans information like in the case


when I draw a molecule without cis/trans or any stereo information. This is needed to feed other programs which


choke on stereo smiles or SDF (yes they exist) but are very specific so JChem is the base development platform and feeds other specialized programs.





So if JChem always adds cis/trans, thats ok, but the Standardizer could allow to get rid of such information.

Code:

trans C\C=C\N1C=CC=C1 cis   C\C=C/N1C=CC=C1 nothing CC=CN1C=CC=C1 unique C\C=C\n1cccc1

The problem here is that the nothing code is recognized


as (E) trans isomer. Instead I would like to have it as "nothing". So how would you store the SMILES code


if you sell a racemic mixture and wiggly bonds are not allowed? I think this is a backwards compatibility issue


and a SMILES issue (how many SMILES codes can I have for one structure?).





Tobias