Extended Smiles with coordinates

Hello,





I store alignment templates as Extended Smiles with co-ordinates.





The user draws the template they wish to align to using Marvin Sketch,


(see Macrocycle.mrv)





Molecule mol = m_MarvinSketchPane.getMol();


String sMol = mol.exportToFormat("cxsmiles:+c");





sMol = CC(C)(C)OC(=O)N[C@H]1CCCCC\C=C/[C@@H]2C[C@]2(CC(=C)[C@@H]2C[C@H](CN2C1=O)OC(=O)N1CC2=CC=CC(F)=C2C1)C(=O)NS(=O)(=O)C1CC1 |r,c:14,39,42,t:37,(-10.46,1.54,;-10.86,.05,;-11.26,-1.44,;-12.19,.82,;-9.53,-.72,;-8.19,.05,;-8.19,1.59,;-7.1,-1.04,;-5.61,-.64,;-4.02,-1.92,;-5.11,-3.01,;-4.71,-4.49,;-3.22,-4.89,;-2.13,-3.8,;-.65,-4.2,;.44,-3.11,;.04,-1.63,;1.58,-1.63,;.81,-.29,;.04,1.04,;-1.5,1.04,;-1.89,-.45,;-2.58,2.13,;-2.11,3.59,;-3.35,4.5,;-4.6,3.59,;-4.12,2.13,;-5.46,1.36,;-6.95,1.76,;-3.35,6.04,;-2.27,7.13,;-2.66,8.62,;-.73,7.13,;-.25,5.66,;1.29,5.66,;2.32,4.52,;3.83,4.84,;4.3,6.3,;3.27,7.45,;4.04,8.78,;1.77,7.13,;.52,8.03,;2.3,.11,;2.3,1.65,;3.64,-.66,;4.97,.11,;4.57,1.59,;6.06,1.2,;6.06,-.98,;5.29,-2.32,;6.83,-2.32,)|





When the Molecule is displayed back to the user, it appears modified (e.g. the chirality is not displayed on the same bonds as before).





mol = MolImporter.importMol(sMol);


marvinSketchPane.setMol(mol);





What can I do to solve this problem?


I am trying to avoid storing the Mol File, or the mrv file.





I have attached the molecule before and after.


I am using Marvin 5.1.1





Thanks


Catherine

The original molecule is Macrocycle.mrv





Thanks


Catherine

It seems to be a bug.


I will examine it ASAP





Andras

Thanks


Catherine

Dear Catherine,





I have examined the problem.


As far as I understand the problem is not that the stereo information of the molecule changes but the wedges are rearranged.


Am I right, is this the case?


If so, than I should say that cxsmiles is not retaining the wedge arrangement. Only the wiggly bonds are preserved.


The reason is the following:


Generally, smiles information does not contain any coordinate information, so fixing wedges to bonds may easily contradict with the given parity specification if coordinates generated during the display process.


So I would suggest you to use different molecule format, which retains this information, like mrv.





Andras

Hello





Yes, you are correct, the chirality has not changed but the wedges have.

Code:

smiles information does not contain any coordinate information, so fixing wedges to bonds may easily contradict with the given parity specification if coordinates generated during the display process.

But I am using Extended Smiles with co-ordinates, so the co-ordinates should not be generated but read from the file ?





Do you mean that the wedges and hash depiction are not stored and created on the fly?





Thanks


Catherine

Hi,

Quote:

But I am using Extended Smiles with co-ordinates, so the co-ordinates should not be generated but read from the file ? Do you mean that the wedges and hash depiction are not stored and created on the fly?

Exactly. The wedge and hash bond information is not stored but created on the fly.





I agree that in case of Extended Smiles if the coordinates are stored it would be possible retain wedge and hash bond information, but we have not think about that until now as the parity information which is chemically equivalent whit this is already stored in the smiles string.





Andras

Hi





We would be very keen if you were to implement it in one of the future versions.





Thanks


Catherine

Hi Catherine,





We decided to implement this feature, but we are full until 5.2, so this can happen only after; probably in 5.2.1.





Best regards,





Akos

Hi Catherine,

We have implemented this feature. It will be available in Marvin 5.2.1, coming out soon.

Andras

Thank you very much, I have just tested it and it is perfect.

Catherine