Aflatoxin Round 5

Case number:699969-2004553
Opened by:jeff101
Opened on:Tuesday, December 12, 2017 - 23:41
Last modified:Saturday, December 16, 2017 - 17:00

What will happen when Puzzle 1455 ends?
Will there be a blog/video/chat giving
us feedback about Rounds 1-4 before
Aflatoxin Round 5 is posted? Will
Round 5 be like Rounds 1 & 3 but with
a 3rd orientation for the Aflatoxin?

(Tue, 12/12/2017 - 23:41  |  4 comments)

Susume's picture
User offline. Last seen 6 days 4 hours ago. Offline
Joined: 10/02/2011

I agree some feedback before another puzzle would be very helpful - I feel like I am working in the dark in terms of what is useful. Is it ok to add only one or two bonds if the shape is good? Do we need to add as many bonds as possible? Do voids beside the ligand make an otherwise promising solution invalid? Should we keep the rest of the protein the same or let it mutate freely (points say let it mutate, but maybe it would be more likely to fold up if we left much of it unchanged)?

Some thoughts on things that would make future aflatoxin puzzles (or any enzyme redesign) easier:

Allow us to have fewer residues than the puzzle starts with. In order to shorten one of the given loops, I had to jam extra residues in somewhere else because it would not allow me to delete unless I first added.

Allow us to reset a selected area of the puzzle to its default without resetting the whole puzzle (I realize this may be hard since the residue numbers can change, but maybe resetting everything from one gray cut to the next could be done).

Joined: 03/05/2015
Groups: Gargleblasters

Also... it may be a moot point due to the possibility of a smaller started puzzle next time, but I noticed that (at least in the Mac clients) if there were inserted segments the puzzle was more prone to crashing on mutate. This seems part of a Mac client problem that extends to other mutable puzzles (the variable length monomers) and if it is consistent across platforms may be a bug worth fixing before the next aflatoxin release.

jeff101's picture
User offline. Last seen 15 hours 20 min ago. Offline
Joined: 04/20/2012
Groups: Go Science

Will Round 5 be like Round 2, where we can import solutions from Rounds 1 & 2?
Will Round 5 be a multi-start puzzle, using a handful of the Foldit Team's
favorite solutions from Rounds 1 & 2 as starting structures to choose from?

bkoep's picture
User offline. Last seen 8 hours 45 min ago. Offline
Joined: 11/15/2012
Groups: Foldit Staff

Good questions! We're going to take a break from aflatoxin for a week or two and give the Siegel lab some time to analyze the results of the first four rounds. Hopefully they will be able to offer you some more specific feedback about design goals and intentions.

We've gotten a lot of great feedback from players so far, and will address as much as we can in the next aflatoxin puzzle. In particular, we may be able to cut down on the size of the puzzle; unfortunately, this will probably prohibit players from loading solutions from the previous puzzles. However, if we see anything striking from the first rounds, we'll definitely include those as starting solutions so the rest of the community can contribute!


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, Amazon, Microsoft, Adobe, Boehringer Ingelheim, RosettaCommons