There is an option to recompute the hydrogens... (RECOMPUTE_HYDROGENS) https://github.com/cdk/cdk/blob/main/tool/smarts/src/main/java/org/openscience/cdk/smirks/SmirksOption.java#L80
Note the explicit [H] is better than setting H1 because unless you specify H0 you may have invalid valence, e.g. [OH-] => [OH] would become. The way the explicit H works is efficient in that it +1 to hydrogen count rather than create an explicit atom. On Mon, 25 Nov 2024 at 00:30, Uli Fechner <u...@pending.ai> wrote: > There is a fair bit of information associated with the original PR: > https://github.com/cdk/cdk/pull/916 > > It's my understanding that hydrogens are not automatically adjusted, that > is, modifying the charge on an atom requires adjusting the hydrogen count. > > In this comment are references to more information regarding the different > matching modes: > https://github.com/cdk/cdk/pull/916#issuecomment-1273164172 > > Best > Uli > > On Mon, Nov 25, 2024 at 4:59 AM Jonas Schaub via Cdk-user < > cdk-user@lists.sourceforge.net> wrote: > >> Hi Egon, >> >> >> >> I also started playing with SMIRKS only a few days ago but I think you >> need to specify the neutral charge and hydrogen saturation in the product >> explicitly. I tried “[O-:1]>>[O;+0;H1:1]” and it produced the output you >> are looking for (O=C(O)C). >> >> >> >> > let's say we have multiple carboxylic acids and amine groups, how would >> one enumerate all possible charge states? Is that possible with just SMIRKS? >> >> >> >> If you use the transform mode “Unique”, you will get as many structures >> returned as there are matches to your SMIRKS in the molecule. So in each >> returned structure, one carboxylic acid would be neutralised. But I guess >> this is only the first step towards what you are looking for. I am actually >> faced with a similar problem right now, i.e. enumerating all possible >> transformation combinations for one molecule and multiple SMIRKS… >> >> >> >> Hope this was of help. >> >> >> >> Kind regards, >> >> Jonas >> >> >> >> *From:* Egon Willighagen <egon.willigha...@gmail.com> >> *Sent:* Saturday, November 23, 2024 6:52 PM >> *To:* CDK users list <cdk-user@lists.sourceforge.net> >> *Subject:* [Cdk-user] SMIRKS in 2.10-SNAPSHOT >> >> >> >> >> Hi John, all, >> >> >> >> I am playing with SMIRKS but it's pretty new to me. >> >> >> >> Transform neutralAcid = Smirks.compile("[O-:1]>>[O:1]"); >> >> IAtomContainer cdkStruct = parser.parseSmiles("CC(=O)[O-]"); >> Iterable<IAtomContainer> iterable = neutralAcid.apply(cdkStruct, >> Transform.Mode.Exclusive); >> for (IAtomContainer neutral : iterable) { >> String neutralSmiles = generator.createSMILES(neutral); >> >> System.out.println(neutralSmiles); >> } >> >> >> >> (I hope I did not make any copy/paste typos) >> >> >> >> Why am I not getting this as output: CC(=O)O ? >> >> >> >> (Actually, let's assume I can get that working, let's say we have >> multiple carboxylic acids and amine groups, how would one enumerate all >> possible charge states? Is that possible with just SMIRKS?) >> >> >> >> Egon >> >> >> >> -- >> >> [NL] WikiPathways in actie voor MetaKids: We zamelen geld in voor >> MetaKids met een actie rond WikiPathways, zie >> https://sr24.wikipathways.org/. Doneer via >> https://www.npo3fm.nl/kominactie/acties/wikipathways-in-actie-voor-metakids >> >> >> >> [EN] WikiPathways in action for MetaKids: We are fundraising for the >> Dutch charity MetaKids by improving metabolic disorder pathways, see >> https://sr24.wikipathways.org/. Donate at >> https://www.npo3fm.nl/kominactie/acties/wikipathways-in-actie-voor-metakids >> >> >> >> -- >> >> E.L. Willighagen >> Department of Translational Genomics >> >> NUTRIM Institute of Nutrition and Translational Research in Metabolism >> Maastricht University >> Blog: https://chem-bla-ics.linkedchemistry.info/ >> Mastodon: https://social.edu.nl/@egonw >> PubList: https://orcid.org/0000-0001-7542-0286 >> _______________________________________________ >> Cdk-user mailing list >> Cdk-user@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/cdk-user >> > _______________________________________________ > Cdk-user mailing list > Cdk-user@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/cdk-user >
_______________________________________________ Cdk-user mailing list Cdk-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/cdk-user
