Closed since over 10 years ago
Intermediate Overall DesignThis is a repost of the trimer from Puzzle 1087, but now the three chains will NOT behave symmetrically! This means that each chain may have a different sequence, and players must design three separate interfaces. Satisfy the Hydrogen Bond Network filter by mutating residues to form HBonds between the Red and Blue polar atoms on the sidechains.
One thing that happened in the recent devprev repost of 1087 (http://fold.it/portal/node/2000825) has also happened in puzzle 1103 once so far:
One recipe I used in both puzzles has ended for unknown reasons in an unusual manner, saying "User cancelled….." each time.
you cannot arrange /strengthen possible H-bonds in the core of the protein, if the backbone (or each chain) is totally locked. After Wiggle, a lot of my H-bonds remain red within the H-bond-filter. An exploration filter would have been better…
why this puzzle type is here - it is substantially artificial - and better suited to other "play modes". I'm not trying to be difficult (I appreciate foldit is Beta) but this has been thro deprev testing - perhaps a scientist will be kind enough to explain why we have this in it's current form? I could go on - puzzle size, lack of tools etc…..but some clear argument could possibly help us players out.
I'm sorry, I don't follow your meaning. Can you elaborate on your concerns? Is this a question of the puzzle's scientific merit, or do the rules of this particular puzzle seem unreasonable?
This Hydrogen Bond Network filter is very new to Foldit—we're still figuring out how useful it is, and parts of it are still in development. The "HBNet" puzzles posted thus far have been designed to mimic Rosetta's capabilities as much as possible. We'd like to compare the ways in which Rosetta algorithms and Foldit players can use the same filter.
I think the question centers on why the HBN puzzles aren't as much fun to work as others. Nothing much to move around and figure out except the AAs… and it seems as though it's all simply a matter of wiggling those AAs around and mutating them in ways that are not as much fun as the puzzles that don't have frozen backbones. It's challenging just because it's a tough puzzle but it's not as much fun.
Hope that helps. Quantifying "fun" can be tricky. I guess the human element doesn't enjoy working "like a computer" as much as might have been expected. (that sounds snarky but isn't at all.) I don't mind difficult, challenging puzzles but having tools that help (things that "light up" areas of interest, color contrast, ability to move and isolate areas that I can't quite reach, etc.). And right now the only tools we have are recipes that are evolving slowly to meet the challenges.
The HBN puzzles also tend to be visually confusing, which is also more difficult for humans to sort out, because the loops and networks are all sort of a mishmash with no way to move them out of the way to see what's going on.
I am also interested in how the social aspect of foldit groups will affect these puzzles over time - and how the foldit developers and engineers look at the ways in which group dynamics come together to improve the folding - are we "teaching the computer" how to do it better, or are we being taught by our interaction with the computer how to do it better? Or is that a cyclical thing. (sorry, early morning musings.)
