Updates from the Foldit Team

Here are some things the Foldit team has been working on lately:

  • jflat06 has just put out a release with a new puzzle type and is working on various bug fixes (electron density crash on Linux, symmetry crash on Windows, and working on stability of the Mac build).
  • bkoep has been busy testing Foldit players' designs in the lab (see image at bottom).
  • scooper is working with some students on developing an interface for playing Foldit on touchscreen devices. He is also working with a student to update tutorial levels.
  • smortier is working on finding outlets for additional outreach opportunities.
  • As a result of the last Science Chat and subsequent player posts, the team is now discussing how we can provide a more "game-like" feel to Foldit. Thank you to those who have already put in suggestions. If you have thoughts, feel free to let us know! You guys are the experts and we value your opinions greatly.

That's it for now, but we'll work hard to keep you in the loop. As an aside, check out this cool article that Foldit was just feature in!

Desktop computer sitting on lab counter, desktop displays graph
Running a circular dichroism experiment with a Foldit player-designed protein. This protein appears to be partially unfolded at room temperature.

UPDATES DISCLAIMER: We want to keep the community apprised of what we are working on, but please note that not everything we spend time on will necessarily come to fruition quickly or even at all. The nature of research is nebulous and often there is no clear answer to a problem. Researchers can work on something for a long time and end up without usable results. Also please keep in mind that we are working on projects that the team has deemed the most appropriate/valuable use of our time and resources based on the information we have at hand. This may not jive with what all players feel is the best for Foldit and while we are certainly willing to listen to your concerns, please understand this may not change our direction. We value, put weight on, and appreciate your feedback nonetheless.

(Tue, 07/25/2017 - 21:11  |  6 comments)
Susume's picture
User offline. Last seen 4 hours 55 min ago. Offline
Joined: 10/02/2011
Thanks for the update!

It's very nice to hear what you all are working on. Thanks!

S0ckrates's picture
User offline. Last seen 14 hours 19 min ago. Offline
Joined: 05/19/2017
Groups: None

I can't wait to see these updates come to fruition! Keep us posted.

Glen B's picture
User offline. Last seen 5 hours 21 min ago. Offline
Joined: 11/25/2014
Groups: None
Rebuild vs Remix

Still confused on why the rebuild tool was removed from the monomer designs when its still available on all other puzzles. The Remix tool can be "less than fun", especially when trying to modify a three segment section and get a "no replacement found", but, add a fourth and you wind up with 94 options. I really miss the "dimer"and "trimer" puzzles.

bkoep's picture
User offline. Last seen 23 hours 35 min ago. Offline
Joined: 11/15/2012
Groups: None
Rebuild is bad for design

I understand the frustration from the point of gameplay experience, but this has been important for improving the stability of Foldit designs.

It turns out that loops are pretty important when designing proteins—just one suboptimal loop can mean the failure of a designed protein. Remix uses a limited library of loops that are fairly common in known protein structures (similarly to the Ideal Loop types, but a little more lenient). The prevalence of these loops suggests that they are inherently stable, and restricting ourselves to just these loops has drastically improved the success rate and stability of Foldit designed proteins. If Remix reports "no replacement found," this means that there is no loop in the library that can connect your two endpoints. Possibly the only way to connect your two endpoints is to use an unstable backbone, and we cannot afford to cut corners in designed proteins.

Rebuild is not very strict about the quality of loops it makes. It simply tries to connect two endpoints in space—even if it requires a completely unrealistic backbone. We tolerate this in structure prediction, because nature is wily and will sometimes find a way to use an unstable loop (e.g. perhaps that loop destabilizes the protein, but is important for the protein's function).

bkoep's picture
User offline. Last seen 23 hours 35 min ago. Offline
Joined: 11/15/2012
Groups: None

We miss the symmetric dimer and trimer puzzles, too! Unfortunately, there's a really nasty bug somewhere in the code that is causing frequent crashes with symmetry puzzles. It's been really difficult to track down, but we're trying some new approaches and will hopefully have a fix soon!

Joined: 06/06/2013
Groups: Gargleblasters
updates greatly appreciated

Thanks. I melt at outdoor temperature in the summer ;)

Get Started: Download
  Windows    OSX    Linux  
(10.7 or later)

Are you new to Foldit? Click here.

Are you a student? Click here.

Are you an educator? Click here.
Only search fold.it
Recommend Foldit
User login
Top New Users

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Microsoft, Adobe, RosettaCommons