Talk:Chemical graph generator

Did you know nomination

The following is an archived discussion of the DYK nomination of the article below. Please do not modify this page. Subsequent comments should be made on the appropriate discussion page (such as this nomination's talk page, the article's talk page or Wikipedia talk:Did you know), unless there is consensus to re-open the discussion at this page. No further edits should be made to this page.

The result was: promoted by Amakuru (talk) 23:10, 16 March 2021 (UTC)

(

• Comment or view
• Article history

)

Chemical graph of the overlapping substructures of caffeine

• ... that one of the first chemical graph generators was developed in the context of NASA's Mariner program to facilitate the automated identification of chemical compounds in search for life on Mars? Sources: "the DENDRAL program contains a structure GENERATOR as its core, abundantly constrained by a set of relevant heuristics. The GENERATOR is built upon a consideration of the conventional structure representation as a topological graph, i.e., the connectivity relations of a set of chemical atoms taken as nodes." "Remarkably, DENDRAL was the first expert system with “outside client” due to Joshua Lederberg's participation in NASA Mariner campaign to search for life on Mars; inter alia, a mass spectrometer and a computer running DENDRAL were planned to be onboard."; "The non-AI motivation, and importance of the overall problem to an outside "client", came from Lederberg's involvement in NASA's Mariner mission and its on-board experiments to look for evidence of life on Mars. A mass spectrometer was part of the on-board instrumentation and Lederberg wanted means of identifying chemical compounds on Mars that were not also found and catalogued in Earth-based libraries. "

• Reviewed: Template:Did you know nominations/El Apóstol

Created by Daniel Mietchen (talk). Self-nominated at 07:32, 29 January 2021 (UTC).

• @Daniel Mietchen: Can you explain the relationship between this page and this article? --evrik ^(talk) 23:38, 7 February 2021 (UTC)

• @Evrik: Sure. This page (on an external wiki) was used as a drafting environment for the Wikipedia article Chemical graph generator. That wiki is operated by PLOS in order to facilitate the creation of PLOS Topic Pages, i.e. articles co-published between them and Wikipedia. The main reason for drafting the articles over there is that the copyright licenses in use at PLOS (CC BY) and Wikipedia (CC BY-SA) are compatible in only one direction, so articles drafted here would not fit in there, but the opposite works. -- Daniel Mietchen (talk) 03:40, 8 February 2021 (UTC)

References

• New enough, long enough, hook interesting and cited. Passes earwig. QPQ Pending. --evrik ^(talk) 05:52, 9 February 2021 (UTC)
• Donating QPQ

@Daniel Mietchen, Evrik, and Edge3: sorry for the late notice on this one, but when I was checked it in the queue it seems like there are large numbers of unreferenced statements in the article. I haven't tagged the statements individually with {{cn}} tags, as almost every paragraph in the body seems to have them. Just as one example:

"Jean-Loup Faulon's structure generator relies on equivalence classes over atoms. Atoms with the same interaction type and element are grouped in the same equivalence class. Rather than extending all atoms in a molecule, one atom from each class is connected with other atoms. Similar to the former generator, Julio Peironcely's structure generator, OMG, takes atoms and substructures as inputs and extends the structures using a breadth-first search method. This tree extension terminates when all the branches reach saturated structures.".

All of that needs citing. Please can you go through and add cites so that this can be re-promoted? Cheers  — Amakuru (talk) 10:17, 26 February 2021 (UTC)

@Evrik and Amakuru: I think I have a solution. Let me explain as this is about odd.

This wiki article is almost entirely a copy of the free and open PLOS article

• Yirik, Mehmet Aziz; Steinbeck, Christoph (5 January 2021). "Chemical graph generators". PLOS Computational Biology. 17 (1): e1008504. Bibcode:2021PLSCB..17E8504Y. doi:10.1371/journal.pcbi.1008504. PMC 7785115. PMID 33400699.

The process for moving such content to Wikipedia is documented at Wikipedia:Journal to wiki publication.

The general idea is that when experts want to present an encyclopedic topic and want it to be free and open, they can write it themselves off wiki, put it through peer review, and have it published. In that workflow anyone can copy it to Wikipedia. It happens that PLOS uses MediaWiki just like Wikipedia, so even the code can be copied over and not just the text.

The original code is in PLOS's wiki at http://topicpageswiki.plos.org/wiki/Chemical_graph_generators, which is useful for wiki but for actual citations the peer reviewed publication is the source.

So to fix the Wikipedia version, I will cite that PLOS published peer reviewed article for every sentence without a citation, because that is all expert peer reviewed content. The original publication does not have citations because these are expert claims, typically of content too obvious for an expert to provide a source. Seem cool? I presume that would pass this.

By the way, I collaborate with Daniel. We organize collaborations between academic journals and Wikipedia. Blue Rasberry (talk) 00:12, 10 March 2021 (UTC)

@Evrik and Amakuru: I cited the original publication throughout the article. Lack of citations is no longer an issue here. Do you see other barriers to advancing the DYK process? Again, I recognize that this is not Wikipedia's typical publishing model so please ask if we should explain more. Blue Rasberry (talk) 21:14, 10 March 2021 (UTC)

• I'm okay with this. --evrik ^(talk) 05:32, 12 March 2021 (UTC)

  @Evrik and Bluerasberry: sorry for the delay coming back to this, this is a situation I've not encountered before actually but all seems fine as noted. Evrik please could you restore a tick to this and I can re-promote it. Cheers  — Amakuru (talk) 20:51, 16 March 2021 (UTC)

• --evrik ^(talk) 21:12, 16 March 2021 (UTC)

Problems

1. The number of citations of ref. 1 is grossly excessive.

2. How is the stoichiometry of the target molecule established?

3. The issue of isomers is not discussed. See butanol for a simple example. What about chirality?

4. With larger molecules they may be more than one chemical compound that are not structural isomers, but have the same stoichiometry.Petergans (talk) 14:41, 21 March 2021 (UTC)

Regarding the first point: citation #1 is this article (as published in a peer-reviewed journal); it's cited a lot because it's the source for all of the content that isn't just summarization of other pre-existing publications. -Bryan Rutherford (talk) 20:04, 21 March 2021 (UTC)

That's not the issue. it's the NUMBER OF TIMES that it is cited. A review does not need to be cited so many times; it would be sufficient to state that is a review and to summarize its content. For specific content it would be preferable to cite the original publication. Petergans (talk) 20:49, 21 March 2021 (UTC)

@Petergans: This is a case where the source cited so much is an original claim by experts through the process documented at Wikipedia:Journal to wiki publication. I assisted in adding all these citations as documented in the "did you know" process above. I also advocate for experts producing content for publication in Wikipedia after getting it peer reviewed in external expert processes. If you have comments about migrating content from journals to wiki, then could you share on that documentation page? That practice is bigger than this one article. Blue Rasberry (talk) 18:23, 5 April 2021 (UTC)

This is an unusual situation in that the journal concerned is an open-access journal, so the contents of the original paper could be copied to WP without breaking copyright law. My view is that nothing was gained by doing so as readers have open access to the original.

With any published paper a single citation is generally sufficient in a WP article. At most I would put a repeat reference once in each major section. As things stand, it has the appearance, rightly or wrongly, of self-glorification of an author of the original publication. Petergans (talk) 19:42, 5 April 2021 (UTC)

Re. agree on principle that even if there is a high quality source which has compatible copyright we should not just copy paste from it (this also applies to articles copied from PD sources such as old editions of Britannica; ...). But whatever. Regarding point 1) I tried fixing it but there's just a lot of it and I am otherwise busy... Cheers, RandomCanadian (talk / contribs) 22:51, 5 April 2021 (UTC)

There's still about 40 of them left. I'll try seeing if I can cut some more. RandomCanadian (talk / contribs) 13:19, 6 April 2021 (UTC)