If the ligand design tool was open when Foldit is closed, it will appear again on the next small molecule design puzzle opened. It's not necessary to select the ligand, or use any icons or hotkeys, the tool is just there when the puzzle opens. (Windows behavior at least.)
On puzzle 2360, the ligand design tool is not supposed to be available. All compounds are supposed to come from the ligand queue or a compound library search.
Players may see the ligand design tool appearing unexpectedly on 2360. Also, players can simply open a previous ligand design puzzle, open the ligand design tool, close Foldit, and restart, opening puzzle 2360.
The ligand design tool seems to work normally on 2360, so nothing is stopping players from designing novel compounds.
The ligand design tool does not appear when changing puzzles. Closing and restarting Foldit is necessary to get the effect.
The image shows the starting compound for 2360 with several atoms converted to fluorine.
As mentioned in my comment to Puzzle 2452, this bug is still present (also in the brand new V34-20240506-85f0ab2ee2-win_x64-devprev).
In addition to what Loci reported above, I found that appearance of the Small Molecule Design panel depends on its state at the time the previous client was closed, not on whether it was visible in the previously loaded puzzle. In other words: I need to close a client with the panel visible, and then open a new client. In that client, I can repeatedly load any puzzle and have the panel.
When I accidentally close the panel (muscle memory :-), Puzzle 2452 does not allow me to re-open it with its button (which isn't shown) or its hotkey ("L" doesn't work). But in that case, it's not necessary to go to another Small Molecule Design puzzle to re-open the panel. All I need to do is re-load the same puzzle in that client, and the panel will be visible (and fully functional) again.
I can confirm I can reproduce this. I think I have it tracked down in the code, and a fix is making its way toward the next devprev.
Please invalidate all score updates coming from clients prior to the latest devprev (35) as they are all susceptible to ligand modification which yields points for useless structures. They may also allow for unduly score increases by replacing existing atoms in the ligand by the same atom (thus retaining scientific value) while optimizing the ligand structure (a side-loophole).
I am currently exploiting this loophole to prove my point (and nudge you into action since it was first reported by Loci october 2023).
I would be very interested to see how many times this was exploited in the past, not only to see where points were unduly credited but also to see what data was scientifically compromised.
Moving forward, score- and ranking-integrity should take a very high priority. Fairness is not the only consideration. The tainted results achieved by those who have used this loophole should not be used to support scholarly work. (GIGO).
The most recent devprev should now have a fix for this issue.
Yes, I noticed, thank you. And that is why I am using an old client to circumvent the scoring system and rig it and I'm not the only one who does that. @beta_helix .Please invalidate all points for the CASP series as the scoring is rigged.
Indeed the most recent devprev (in my case, V35-20240517-3a0ae6a15b-win_x64-devprev) offers a fix for this issue. That is to say, I tried to bring up the Small Molecule Design panel in one of the CASP puzzles, and I did not succeed.
But I fully agree with @"Bletchley Park" (in the two comments he made above) that this doesn't mean that "cheating" is no longer possible.
I wrote "cheating" in quotation marks, because I think there are two possibilities. You can use a trick (opening an old puzzle, in release V34 or earlier) to make the Small Molecule Design panel appear. But it is also very well possible that a player simply had the tool at their disposal because their screen is large enough and they just never closed the panel.
In any case, the value for the CASP competition would be zero.
See also my suggestion in the comments to Puzzle 2452.
