Deleting H atoms with stereobonds affects other stereobonds

User 870ab5b546

22-05-2005 03:26:22

If you delete the H atoms in the structure below with the erase tool, you will see that remote stereocenters lose their stereobonds.





This behavior occurs in MarvinSketch 3.5.5, 3.5.6, and 4.0alpha6. It occurs for many different molecules in Fischer projections like the one below, even when one of the OH groups is replaced with NH2, but I don't know whether it occurs for other compounds.





Code:






  Marvin  05210523192D





 18 17  0  0  0  0            999 V2000


   -4.1250   -1.0312    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0


   -3.3281   -1.2448    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0


   -3.3000   -0.2062    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0


   -4.9500    0.6188    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0


   -4.9500    1.4438    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0


   -4.1250    1.4438    0.0000 C   0  0  1  0  0  0  0  0  0  0  0  0


   -3.3000    1.4438    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0


   -4.1250    0.6188    0.0000 C   0  0  2  0  0  0  0  0  0  0  0  0


   -3.3000    0.6188    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0


   -4.1250   -0.2062    0.0000 C   0  0  1  0  0  0  0  0  0  0  0  0


   -4.9500   -0.2062    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0


   -3.3000    2.2688    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0


   -4.9500    3.0938    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0


   -4.1250    3.9188    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0


   -3.4105    4.3313    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0


   -4.1250    3.0938    0.0000 C   0  0  3  0  0  0  0  0  0  0  0  0


   -4.1250    2.2688    0.0000 C   0  0  1  0  0  0  0  0  0  0  0  0


   -4.9500    2.2688    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0


 17  6  1  0  0  0  0


  6  8  1  0  0  0  0


  8 10  1  0  0  0  0


 10  1  1  0  0  0  0


  1  2  1  0  0  0  0


 10  3  1  1  0  0  0


  8  4  1  1  0  0  0


  6  5  1  1  0  0  0


  6  7  1  1  0  0  0


  8  9  1  1  0  0  0


 10 11  1  1  0  0  0


 17 16  1  0  0  0  0


 16 14  1  0  0  0  0


 17 12  1  1  0  0  0


 16 13  1  0  0  0  0


 14 15  2  0  0  0  0


 17 18  1  1  0  0  0


M  END


ChemAxon 7c2d26e5cf

23-05-2005 19:18:04

We will check the issue.

ChemAxon 25dcd765a3

24-05-2005 11:15:20

To determine chirality we use the determinant.


The center atom is translated to the origin and the connected bonds are treated as vectors.


The determinant is calculated for these vectors.


In this molecule if you delete a H a mathematically degenarate case appears as one center will have 3 ligands from which 2 is in the same line as the center atom.


We try to overcome on this problem soon...


Thank you for the report.

User 870ab5b546

24-05-2005 11:55:05

Hm, interesting.





The reason I make two of the bonds colinear is to make the structure resemble a Fischer projection, which for historical reasons is still used in textbooks.





Perhaps in the case where three bonds are coplanar and at 90° from one another, you can temporarily add the fourth bond pointing to the fourth compass point before computing the determinant.





The other possibility is just to leave the other bonds alone when the fourth bond is deleted.

ChemAxon 25dcd765a3

24-05-2005 11:56:05

I solved the problem. The new release will contain the fix.


Thank you fro the report of this mathematically interesting case.