New puzzles posted in devprev

Started by beta_helix

jflat06 Staff Lv 1

This puzzle uses some symmetric scoring options which we have not used before. Try it out and see if there's anything odd about the behavior of the game's tools.

tamirh Lv 1

This puzzle uses a new feature called Layer Design. It will restrict what residues can mutate to based on its position in the protein. Please try it out and see if you find anything odd about it. Thanks!

jflat06 Staff Lv 1

What we'd like to test with this puzzle is whether anyone is able to move the locked segment that is right next to the hub. It should not be able to move, and if it does, it's a bug. Please let us know if you manage to do it!

tamirh Lv 1

http://fold.it/portal/node/994303

This puzzle features a new filter that we have been working on that should help iron out some of the issues we've been seeing with design puzzles. The filter will identify two main regions (the core and the surface) of the protein. In this puzzle, you are required to have 30% of your residues inside the core of the structure. What this means is that you should not be able to make structures that have no 'inner' region.

What we'd like to test with this puzzle is whether the filter correctly identifies your protein core and if the filter makes sense while you are doing your folding. Proceed with the puzzle as you would fold a symmetric design puzzle. Let us know any comments or bugs you have on it.

Thanks

jflat06 Staff Lv 1

This is a semi-repost of an earlier beta puzzle that had issues:

What we'd like to test with this puzzle is whether anyone is able to move the locked segment that is right next to the hub. It should not be able to move, and if it does, it's a bug. Please let us know if you manage to do it!

beta_helix Staff Lv 1

But we want to make sure it doesn't have as many issues at the first time we posted it (http://fold.it/portal/node/993431).

Please help us by testing the new puzzle out in devprev, "Sepsis Testing": http://fold.it/portal/node/994570

Please report any bugs in the puzzle comments or feel free to open up a new feedback about it.

Thank you very much for beta-testing, as we much rather catch these issues before these design puzzles are posted!

jflat06 Staff Lv 1

This is another iteration of the Dimer Hub. This time, we've added a bonus for forming an Hbond to an atom on the hub.

What we'd like tested is the Hbond bonus - make sure you get the bonus if and only if you have at least one hbond to the highlighted atom (you can highlight the atom with 'show').

tamirh Lv 1

These puzzles are testing weights for the envsmooth (smoothed environment) score term.

What this term does is look at the environment around each amino acid and determine if it is statistically similar to the environment found in similar amino acids in native proteins. In earlier design puzzles, this score term was not used. In puzzle 695 we added this score term and weighted it very highly (at 3.0). We are now testing out smaller values for this score term. Ideally we want to use the highest value for the term possible that does not interrupt your play styles and enjoyment. Please test out the differences you see with shake/wiggle/mutate/running your scripts and using your general techniques you've created for symmetric design puzzles.

You will be able to load in your solutions from 695, and also load back/forth between the 4 different devprev puzzles that are testing envsmooth.

One other thing to keep in mind is how puzzle http://fold.it/portal/node/994886 reacts differently also. There are more variables that are changed in this version, not just envsmooth.

Thanks for all the help!

List of puzzles:
http://fold.it/portal/node/994883
http://fold.it/portal/node/994884
http://fold.it/portal/node/994885
http://fold.it/portal/node/994886

jflat06 Staff Lv 1

This is a test for a new puzzle option - centroid mode. In centroid mode, sidchains are represented as a single large atom. This means no shaking since there's only one position. It also means the puzzle should be quite a bit faster, letting us run larger puzzles.

We're interested in testing for bugs, as well as thoughts on performance. Your feedback would be appreciated!