The SMILES I sent you works fine for me with the same version: https://colab.research.google.com/drive/19OZtX8IqICZQ4B2jLpr02owkSLc0FeZ6?usp=sharing
However, the alkali and alkaline earth metals do not behave as I would expect (as shown in the Colab notebook). This looks like a bug to me and I suggest filing a GitHub issue Best regards, Jan On 22 Mar 2022, at 15.26, Marco Stenta <marco.ste...@gmail.com<mailto:marco.ste...@gmail.com>> wrote: Thanks, Jan, the dative bond works in a number of cases with other metals. (rdkit 2021.9.5) This one works fine: rdmol = Chem.MolFromSmiles('[O-]->[Fe+2]<-[O-]', sanitize=True) # case 1 coordinate bonds assert rdmol is not None this one with Mg divalent does not rdmol = Chem.MolFromSmiles('[O-]->[Mg+2]<-[O-]', sanitize=True) # case 1 coordinate bonds assert rdmol is not None can you read in the smiles you sent me? I can't I am doing anything wrong here? cheers, m rdmol = Chem.MolFromSmiles('C1CO->[Fe+2](O)(<-OC1)(<-O)(<-O)(<-O))=O', sanitize=True) # case 1 coordinate bonds assert rdmol is not None print(rdmol) Il giorno mar 22 mar 2022 alle ore 15:13 Jan Halborg Jensen <jhjen...@chem.ku.dk<mailto:jhjen...@chem.ku.dk>> ha scritto: Hi Marco You can define dative bonds like this: C1CO->[Fe+2](O)(<-OC1)(<-O)(<-O)(<-O) Best regards, Jan On 22 Mar 2022, at 15.07, Marco Stenta <marco.ste...@gmail.com<mailto:marco.ste...@gmail.com>> wrote: You don't often get email from marco.ste...@gmail.com<mailto:marco.ste...@gmail.com>. Learn why this is important<http://aka.ms/LearnAboutSenderIdentification> Dear RDKitters, I am struggling with working organometals and coordination complexes. with a small team, we are creating a series of recommendations to draw correctly organometals (catalyst, complexes, etc), so that we can use them in our chemoinformatics pipeline. I know it is a horrible mess out there, but we are trying to achieve some consistency, rather than full correctness. I guess there is something with the accepted valence for Mg. Now the purpose of this is to have a smiles representation of these metal complexes that are not fragmented (with the dot) so that I keep the notion of bond where there is (or I believe) one for instance, for the edta complex this smiles works fine: [Na+].[Na+].[Mg++].[O-]C(=O)CN(CCN(CC([O-])=O)CC([O-])=O)CC([O-])=O but I would really distinguish the fact that there is a bond between the [O-]/N and the [Mg++ ] while there is none with the two [Na+] <image.png> I can read it in without sanitization, but it fails for everything I nato to do after with the molecules. in theory, dative bonds should not affect the valence of receiving atoms, right? any suggestion for reading in the enclosed v3000 molfile and keeping the bonding info? or sharing what you are doing with metals Thanks a lot in advance, kind regards Marco f1 = 'edta_case.mol' rdmol = Chem.MolFromMolFile(f1, sanitize=True) print(rdmol) Chem.MolToSmiles(rdmol) the rdmol is None in my case <edta_case.mol>_______________________________________________ Rdkit-discuss mailing list Rdkit-discuss@lists.sourceforge.net<mailto:Rdkit-discuss@lists.sourceforge.net> https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.sourceforge.net%2Flists%2Flistinfo%2Frdkit-discuss&data=04%7C01%7Cjhjensen%40chem.ku.dk%7C9b36a1f64d3f4a5a243808da0c0dae6b%7Ca3927f91cda14696af898c9f1ceffa91%7C0%7C0%7C637835550154504697%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=E2VK%2BP1lGqICkgnNtQPHpmEDUuD0qV6VbD0FZNt7w5I%3D&reserved=0<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.sourceforge.net%2Flists%2Flistinfo%2Frdkit-discuss&data=04%7C01%7Cjhjensen%40chem.ku.dk%7Cd425f328d6fb4b92eb5e08da0c10033a%7Ca3927f91cda14696af898c9f1ceffa91%7C0%7C0%7C637835560155987046%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=2aw1SMBTere6vDXsJv2Ae3hfrQvgmUI6ZX29LlkuJpI%3D&reserved=0>
_______________________________________________ Rdkit-discuss mailing list Rdkit-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
