From nbehrnd at yahoo.com Thu Apr 7 10:33:54 2022 From: nbehrnd at yahoo.com (Norwid Behrnd) Date: Thu, 7 Apr 2022 09:33:54 +0200 Subject: [Cod-bugs] InChI days, brief note References: <20220407093354.70ba987b.ref@carnot2> Message-ID: <20220407093354.70ba987b@carnot2> Dear Andrius, earlier, you mentioned interest for (some of) the topics of the InChI Days. By the profiles seen in the zoom meeting on Tuesday/Wednesday, I assume you did not participate because, as already indicated by you in advance, potential overlap of InChI trust's schedule set, and your agenda. Some topics might be of interest for the COD (/vide infra/), thus is included in cc of this email. To keep you in the loop, and to allow a follow up when more convenient for you, I can report there was a wide range of topics discussed. Though there are no video recordings, nor minutes/proceedings, the organizers want to share the speakers' presentations as .pdf; possibly within the next couple of days. As a glimpse, I attach a collection of screen photos taken as a .pdf to this email.* Depending on the topic, the intensity of discussions varied considerably; tautomers and the anticipated contest ?precision FDA? to improve/to foster a more reliable identification and description of structures (especially, but not limited to) tuples of tautomers was an example of the much more lively ones. Speaking of the ?precision FDA contest?, this is going to be /like/ the CCDC blind tests; join and bring your methods, a bench mark of approaches, potential co-authorship of publications. (Different to the CCDC blind tests, though, the invite ?you may bring your own typical/difficult data/structures?.) Since the CCDC's data file was accessed e.g., in Mark's survey about reliable identification of tuples of tautomers, and taking into account CCDC's license policy I boil down to ?of course you may search in our data base, but the systematic extraction of entries aiming e.g., a parametrization of a molecular force field may be disallowed by the CCDC?, it is my speculation a participation by the COD could be of mutual interest and benefit for the contest in preparation, and the visibility of the COD. This is why I include the COD helpdesk with cc of this email. InChI trust organized multiple working groups. Thus, beside the ?InChI core application? to assign InChI strings to e.g., .sdf files is only one among the activities visible to the interested. There are extensions either in preparation, or ongoing improvement e.g., about artificial polymers, ?large molecules? (to address DNA/RNA, proteins), mixtures, reactions, nanoparticles, Markush formulae. Some of them mirror their programs on GitHub, which Gerd Blanke presenting the InChI core application equally anticipates as ?a? .or. ?the? future avenue for InChI programs in general. In the course to yield InChI version 1.07, the underlying code should become more modular to ease modernization and implementation of newly requested features, enable (I think, the intent was ?to deploy a wider range of?) automated unit tests (I speculate, similar to the checks building/using OpenBabel when filing a pull request on GitHub) for quality control of the programs. He explicitly mentioned difficulties to identify potential contributors with insight in programming (C/C++), building these tests .and. relevant background in chemistry. This is an additional reason for me to forward this information to you, and the COD, to facilitate to get in touch with Gerd Blanke/InChI trust just in case you recognize an overlap in interests .and. potentially fruitful complementary areas in expertise. With constraints of the meeting, there was too little time left after Gerd Blanke's presentation ?InChI future? (the very last session of InChI days) to discus how InChI might move to a license scheme more compatible for FOSS/perhaps adopt again a GPL license instead of the bespoke license limiting e.g., Debian/DebiChem to version 1.03 (by June 2010) while the trust distributes an executable in version 1.06 (by December 2020). I speculate if e.g., InChI and OpenBabel would join forces, maintainers of OpenBabel would suggest the adoption of a license easier to recognize/more welcoming to contributions. With best regards, Norwid *) .png joined by ImageMagick, there is no searchable text layer. ---- 8>< ---- file(s) attached ---- ---- 2022-04-06_InChIDays2022_Tuesday.pdf ---- 8>< ---- ---- -------------- next part -------------- A non-text attachment was scrubbed... Name: 2022-04-06_InChIDays2022_Tuesday.pdf Type: application/pdf Size: 4185523 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 833 bytes Desc: OpenPGP digital signature URL: