Dear Janusz, This isn't currently possible. The most straightforward way I could think to implement it (maybe someone else has a better idea?) would be to allow the molecule's RingInfo data structure to be edited so that you could, for example, tell it to "forget" that rings above a certain size exist.
This would be relatively straightforward to do and I could imagine that functionality being useful in other places as well. -greg On Sun, Nov 15, 2015 at 12:08 AM, Janusz Petkowski <[email protected]> wrote: > One other question about MCS, in addition to my previous one on > hybridization: > > In the RDKit documentation in the Maximum Common Substructure (MCS) > section it is mentioned that one can restrict mapping linear fragments on > to rings using two methods: ringMatchesRingOnly and completeRingsOnly. > > It is an extremely useful method but is there a possibility to restrict > execution of this method by defining the size of the rings for which ring > bonds will match only ring bonds in a given molecule? But at the same time, > for rings of a certain size (let's say for rings that have below 8 atoms) > the function is still executed? I am trying to avoid the problem of not > finding linear fragments in the macrocycles structures. I want linear > fragments to be matched in macrocycles but not in rings of smaller > ("regular") size, all within the same molecule of course. Is there a way to > do that? > > Just as an illustration of the problem: > > Is it possible to find ClCC=C fragment in CC(F)C1CC\C=C(C)\C(Cl)C2C=CC( > Br)CC2[C@H](C)[C@]2(O)OC3=C(C)C(=O)C(O)=C([C@@H]4O[C@H]1[C@H](C)[C@H > ](O)[C@H]4C)C3=C2 > but at the same time avoid finding CC(Br)C=C in it, by using the > ringMatchesRingOnly and completeRingsOnly methods? > With current implementation of ringMatchesRingOnly and completeRingsOnly > methods they treat all the rings the same way, no matter the size. > If not, how would one do that? > > Thanks a lot for your help! > > Have a great weekend! > > Best regards, > > Janusz Petkowski > > > ------------------------------------------------------------------------------ > > _______________________________________________ > Rdkit-discuss mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/rdkit-discuss > >
------------------------------------------------------------------------------
_______________________________________________ Rdkit-discuss mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
