Greg, Thanks.
Yingfeng On Fri, Aug 28, 2015 at 6:13 AM, Greg Landrum <[email protected]> wrote: > Well, here's a way to find the rings where all of the bonds are conjugated: > > In [1]: from rdkit import Chem > > In [2]: m = Chem.MolFromSmiles('C1CC1CC(=O)c1nnc2cc(-c3ccc(OC)cc3)nn2c1CC') > > In [3]: ri = m.GetRingInfo() > > In [4]: brs = ri.BondRings() > > In [5]: conjugatedRings = [] > > In [6]: for i,br in enumerate(brs): > conj=True > for b in br: > if not m.GetBondWithIdx(b).GetIsConjugated(): > conj=False > break > if conj: > conjugatedRings.append(i) > ...: > > > Then, using those indices, you can find the lists of atoms that are in > each of those conjugated rings: > > In [7]: ars = ri.AtomRings() > In [8]: conjugatedAtomRings = [ars[x] for x in conjugatedRings] > > In [9]: len(ars) > Out[9]: 4 > > In [10]: len(conjugatedAtomRings) > Out[10]: 3 > > In [11]: conjugatedAtomRings > Out[11]: [(7, 8, 9, 21, 22, 6), (10, 9, 21, 20, 11), (13, 14, 15, 18, 19, > 12)] > > > From this point it's a matter of deciding how you want to count electrons > from each atom. The approach used in the RDKit C++ code is described, more > or less accurately, here: > http://rdkit.org/docs/RDKit_Book.html#aromaticity > > -greg > > > On Thu, Aug 27, 2015 at 5:36 PM, Yingfeng Wang <[email protected]> wrote: > >> Greg, >> >> Thanks. I believe identifying anti-aromatic bonds even just with simple >> rings will be very helpful to my project. Could you please give me some >> links or sample codes about identifying all the conjugated simple rings and >> do an electron count? Would you also like to share me some relevant rules >> you think could be reasonable? >> >> By the way, if there is a hard way to identify all anti-aromatic bonds, >> could you also please give me some hints? If fortunately I can implement >> it, I will share it by posting my code in this mailing list. >> >> >> Thanks. >> >> Yingfeng >> >> On Thu, Aug 27, 2015 at 10:08 AM, Greg Landrum <[email protected]> >> wrote: >> >>> Hi Yingfeng, >>> >>> Non-aromatic is not a problem, but there's no easy way I can think of >>> off the top of my head for identifying anti-aromatic systems without making >>> C++ code changes. >>> >>> If you were willing to limit yourself to simple rings, you could write >>> some python code to identify all the conjugated simple rings and do an >>> electron count according to whatever rules you think are relevant. >>> >>> -greg >>> >>> >>> On Thu, Aug 27, 2015 at 1:52 PM, Yingfeng Wang <[email protected]> >>> wrote: >>> >>>> Could anyone please share your experience with identifying non-aromatic >>>> bonds using RDKit? I would greatly appreciate your help. >>>> >>>> Yingfeng >>>> >>>> >>>> On Mon, Aug 24, 2015 at 9:42 PM, Yingfeng Wang <[email protected]> >>>> wrote: >>>> >>>>> Sorry, my previous email has a typo. >>>>> >>>>> According to the following link, >>>>> >>>>> http://www.rdkit.org/docs/RDKit_Book.html >>>>> >>>>> I get sample code for identifying aromatic bonds like >>>>> >>>>> >>> from rdkit import Chem >>>>> >>> m=Chem.MolFromSmiles('O=C1C=CC(=O)C2=C1OC=CO2') >>>>> >>> m.GetBondBetweenAtoms(6,7).GetIsAromatic() >>>>> >>>>> Is there a way to identify anti-aromatic bonds using Rdkit? >>>>> >>>>> Thanks. >>>>> >>>>> Yingfeng >>>>> >>>>> On Mon, Aug 24, 2015 at 9:38 PM, Yingfeng Wang <[email protected]> >>>>> wrote: >>>>> >>>>>> According to the following link, >>>>>> >>>>>> http://www.rdkit.org/docs/RDKit_Book.html >>>>>> >>>>>> I get sample code for identifying identify anti-aromatic bonds like >>>>>> >>>>>> >>> from rdkit import Chem >>>>>> >>> m=Chem.MolFromSmiles('O=C1C=CC(=O)C2=C1OC=CO2') >>>>>> >>> m.GetBondBetweenAtoms(6,7).GetIsAromatic() >>>>>> >>>>>> Is there a way to identify anti-aromatic bonds using Rdkit? >>>>>> >>>>>> Thanks. >>>>>> >>>>>> Yingfeng >>>>>> >>>>> >>>>> >>>> >>>> >>>> ------------------------------------------------------------------------------ >>>> >>>> _______________________________________________ >>>> 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
