I had a pretty busy week. Thank you for your answer, and forgive my delay. Thank you for your answer; it actually help to partially tackle the problem: there was a problem in the SMARTS I wrote (even if I cannot "see" it)
Since it is possible to transform groups for tautomers, it should be possible to do the same thing for resonant radicals. (from the "computer" point of view I am simply transforming a group into another to prevent name collision) Here comes the question: is it "normal" that in the transformation code I cannot specify either the explicit H atoms? Where should If I try with vinoxy radical C=C[O] and apply the following transformation TRANSFORM [C:1](-[OH0:2])(=[CH2:3]) >> [C:1](=[OH0:2])(-[CH2:3]) c:\Tests>obabel -:C=C[O] -h --mesomers -osmi CC=[O] (wrong #of H on the "left" carbon) Same problem if I try the "opposite" transformation (from keto to enol): TRANSFORM [C:1](=[OH0:2])(-[CH2:3]) >> [C:1](-[OH0:2])(=[CH2:3]) c:\Tests>obabel -:[CH2]C=O -h --mesomers -osmi [CH2]=CO (wrong #of H on the oxygen) Note that the transformation actually occurred: the single and double bond swap. I tryied to understand what is going on debugging the code (and I report here what I found because it might help some of the programmers) As you can imagine, I just scratched the surface. Still, I quite don't understand the following: The struct Pattern contains a boolean variable named hasExplicitH. which effects the behaviour of the Match function. This one, a part during allocation, is changed only on row 1509 of function OBSmartsPattern::SMARTSParser in parsmart.cpp. In particular is set to true only if somewere in the string a "[H]" is found. I cannot understand why. Any hint? it seems to me, that [CH2] in smarts means "an aliphatic C with 2 H attached " and that those 2 H should be explicit. ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure _______________________________________________ OpenBabel-discuss mailing list OpenBabel-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
