Hi Yong, I wanted to say that STOUT only generates correct IUPAC names in 90% of the cases. Like many deep-learning-based applications, it dreams up the rest. Those wrong ones will usually still be relatively close to the correct ones, so it depends on your application whether STOUT is helpful for you. If, for example, you are a chemist (which you aren’t, I read this :), but let’s assume you are) and you want to name a few compounds for a paper, then STOUT is helpful, and you can correct its occasional mistakes and still have a much easier life. If, however, you want to name 10k compounds in your database, then you cannot live with <100% naming success. If you need 100%, you need to switch to an algorithmic tool like Marvin's name generator, referenced in our paper. I hope that makes sense.
Kind regards, Chris — Prof. Dr. Christoph Steinbeck Vice President for Digitalisation of the Friedrich-Schiller-University Jena Analytical Chemistry - Cheminformatics and Chemometrics Friedrich-Schiller-University Jena, Germany Phone Team Assistant: +49-3641-948171 http://cheminf.uni-jena.de http://orcid.org/0000-0001-6966-0814 What is man but that lofty spirit - that sense of enterprise. ... Kirk, "I, Mudd," stardate 4513.3.. > On 28. Jul 2023, at 17:36, Yong Gao <y...@blueprintmedicines.com> wrote: > > Thank you, Chris. > Tried a couple of smiles strings with STOUT. They seem to work. I’m not a > chemist, but does non-deterministic suggest not unique? > Yong From: Christoph Steinbeck <christoph.steinb...@uni-jena.de> > Date: Friday, July 28, 2023 at 4:19 AM > To: Yong Gao <y...@blueprintmedicines.com> > Cc: "cdk-user@lists.sourceforge.net" <cdk-user@lists.sourceforge.net> > Subject: Re: [Cdk-user] IUPAC name generation > [EXTERNAL SENDER] > You can try our STOUT > https://github.com/Kohulan/Smiles-TO-iUpac-Translator > https://jcheminf.biomedcentral.com/articles/10.1186/s13321-021-00512-4 > but be aware that this deep learning tool is not deterministic and makes > occasional mistakes. > If your structures are not too complicated, you can use Egon‘s tip of cross > checking with OPSIN. > If, of course, you want to name large amounts of structures in an > unsupervised manner, you need a deterministic tool. > Kind regards, > Chris > — > Prof. Dr. Christoph Steinbeck > Vice President for Digitalisation of the Friedrich-Schiller-University Jena > > Analytical Chemistry - Cheminformatics and Chemometrics > Friedrich-Schiller-University Jena, Germany > Phone Secretariat: +49-3641-948171 > http://cheminf.uni-jena.de > http://orcid.org/0000-0001-6966-0814 > > What is man but that lofty spirit - that sense of enterprise. > ... Kirk, "I, Mudd," stardate 4513.3.. > > > Am 27.07.2023 um 17:33 schrieb Yong Gao <y...@blueprintmedicines.com>: > Hi, > Has anyone successfully generated IUPAC names from a smiles string? I see > some code in the legacy module, but did not see a way to do it. Also, any > suggestions for doing this with some other open source software? > Thanks, > Yong > CONFIDENTIALITY NOTICE: This electronic mail transmission may contain > privileged, confidential and/or sensitive information and is intended only > for the review of the party to whom it is addressed and for the stated > purpose. Unauthorized use or disclosure is strictly prohibited. If you have > received this transmission in error, please notify > eupriv...@blueprintmedicines.com if you're in the EU or > priv...@blueprintmedicines.com for all other locations. Then immediately > delete the transmission without reading its contents. > _______________________________________________ > Cdk-user mailing list > Cdk-user@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/cdk-user > > CONFIDENTIALITY NOTICE: This electronic mail transmission may contain > privileged, confidential and/or sensitive information and is intended only > for the review of the party to whom it is addressed and for the stated > purpose. Unauthorized use or disclosure is strictly prohibited. If you have > received this transmission in error, please notify > eupriv...@blueprintmedicines.com if you're in the EU or > priv...@blueprintmedicines.com for all other locations. Then immediately > delete the transmission without reading its contents. _______________________________________________ Cdk-user mailing list Cdk-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/cdk-user
