Thanks! My plan as of now is to expand pharmacodynamics sections for more obscure psychedelic substances based on modern studies. I will do my best to adhere to corrections you have made for the future edits:) Wolren (talk) 18:20, 16 October 2025 (UTC)Reply
That sounds great, I look forward to the expansions! It's nice to see someone else making solid contributions in this area. I have to say, I really burnt out on those activity tables after a while. But there are still many more to be done. – AlyInWikiWonderland (talk, contribs) 22:05, 16 October 2025 (UTC)Reply
Latest comment: 6 months ago6 comments3 people in discussion
Hi again. Your 3D ball-and-stick structures are fantastic, thank you so much for doing all of them. I'm sure you gleaned as much that I felt that way from me spamming the "Thanks" button on your edits, but I thought I'd say it here too.:-)
I saw your WikiMolGen project also and it's very impressive. You should consider posting it on Wiki—maybe it could be added to MOS:CSDG. Re: the web app, I was wondering: what settings should be used to generate properly standardized 3D ball-and-stick structure images? E.g., color template, settings template, and so on? I figured out that you probably want 180° Y rotation for tryptamines. There are some missing 3D structures that I'd like to generate and add as well but I want to make sure I do it right. – AlyInWikiWonderland (talk, contribs) 01:39, 29 November 2025 (UTC)Reply
I've made it spontaneously as I was unable to find a similar project, and was supposed to create something using python for university. I think it's nice to have a tool anyone can use instead of having to dive into paid chemistry software or programming. That way the process becomes trivial and more articles can have both of the graphics.
I have no idea which settings for 3D are ideal here. I've just messed around until I got something that could be considered decently visually pleasing. If this tool can become widespread in usage then I believe there is a need for some sort of mutual agreement coming from pharmacology and chemistry wikiprojects on this topic.
About color template the current code doesn't apply any of the example (and sloppy) templates so it defaults to this color ramp. Everything could be easily adopted to custom version of Jmol scheme if needed.
When it comes to rotation the needs vary. For the structures I've generated I mostly used 0 or 180 values for X and Y in various combinations. 2C-X requires Y and Z set to 180. Using the same settings doesn't guarantee 1:1 actual orientation of the compound, at least not with the current setup.
Yes, I definitely agree. It often surprises me that we don't have good free and web-based tools for seemingly simple things like this. Re: mutual agreement from pharm and chem WikiProjects, I think it'd just be WikiProject Chemicals really. And I think it could just be posted and provided as an option, with adoption happening organically. I think the only real thing would be closely following WikiProject Chemicals's structure drawing specifications. So some settings tweaks basically. Good to know, thanks for all the answers. I guess I can just try making some structures that look reasonably good and using those. That's also good to know re: the amine orientation. Maybe I should mostly hold off for now actually considering that. – AlyInWikiWonderland (talk, contribs) 22:53, 30 November 2025 (UTC)Reply
Update: For 3D I'm actually using some custom colors for the main elements for better visual clarity over default ones. I've improved 2D now to use ACS function on default, unlike the previous solution this makes the padding around the structure normalized and should be compliant with MOS:CSDG. The visible size of the structures on the site is also normalized currently. I've also made it so that amines rotate more accordingly.
Unfortunately due to RDKit limitations the more complex compounds don't render perfectly, it has a bunch of strange tendencies when it comes to positioning. Some issues could probably be solved by tinkering and exposing more options on the web, but not every. I'm afraid that making it work perfectly universally would require either a huge amount of logic and/or building on top of the module itself using C++. Time and skill required for those exceed my current capabilities. Unless someone with more experience working with this module helps, or I unexpectedly find some solution, 2D generation should probably be reserved to simple compounds.
I've also added protein generation option but it's really preliminary for now.
Overall I'm planning to fix the template handling, add more options, fix some bugs, make a more thoughtful color palette, post about it in some places by the end of year and call it a day for some time. Wσlrεη⎇ (talk) 21:13, 10 December 2025 (UTC)Reply
Latest comment: 8 months ago2 comments2 people in discussion
Hi there. Thank you for adding the 3D images, but a lot of your 2D images are poor quality and need to be reverted - the atom letters (N, O, S etc) are too large and overlap the lines which show the bonds. Please make sure when changing images that the new ones are not inferior to the one you are replacing. Meodipt (talk) 05:53, 3 December 2025 (UTC)Reply
Thanks for the feedback. I've found a way to generate ACS compliant images now so for the futurre: too large font size, too low padding between atom letters and bonds, too dense stereochemistry ticks, non-standardised margin issues are fixed. Is there any other specific general issue with the images? Wσlrεη⎇ (talk) 21:21, 10 December 2025 (UTC)Reply
@AlyInWikiWonderland I have fixed the missing rotation in the WikiMolGen and introduced a bunch of new features, as well as data fetching strategies. Template logic was also reworked and renamed to presets (one per mode, color palettes in a separate category down). Sorry in advance if that broke something you have had previously prepared. I think I am closer to settling on the mental model of this right now and things should slowly become more stable from here. You can report any inconvenience as I will be able to actively work on projects in this and upcoming months. Wσlrεη⎇ (talk) 23:31, 21 June 2026 (UTC)Reply
