Foldit Puzzles
Play puzzles to help scientific research and compete with other players. New puzzles are posted every week.
-
We are giving you CASP9 Puzzle 1 again, but this time with the native structure as a guide. Use the alignment tool and try to match the native as much as possible. After threading, make sure to press G to Align the Guide.
-
We are giving you CASP9 Puzzle 1 again, but this time with the native structure as a guide. Use the alignment tool and try to match the native as much as possible. After threading, make sure to press G to Align the Guide. For players with fewer than 150 global points.
-
Is that...?! Yes, it is! It's TNT! We've got a very energetic, high profile molecule for you guys to bind. Sure, we already have antibodies that bind TNT, but they're very expensive to make. Wouldn't it be better if we just designed an easy to express protein that bound TNT? We could use it to make cheap detectors and sensors so we can look for TNT contamination or explosives! Brilliant! Remember, there is more than one way to play these design puzzles since we look at all the designs and the best scoring design may not necessarily be the most useful/promising. Check out the puzzle comments for hints!
-
Our last CASP9 Refinement puzzle is the same target from 342: CASP9 Puzzle 34. This is the one of the most challenging refinement targets. There are big deviations from the native structure in the areas around residues 10, 20 and 50. These regions are denoted in note mode (press 3 to enter Note Mode), and your score will not register until you have sufficiently refined this area. Please see the puzzle comments for more info as we will post a different version of this puzzle.
-
This is another starting point for our last CASP9 Refinement puzzle (same as puzzle 353). Since this is the one of the most challenging refinement targets, with big deviations from the native structure in the areas around residues 10, 20 and 50 (these regions are denoted in note mode; press 3 to enter Note Mode), we are giving you a starting model that has been threaded to the model given to us by CASP except for these areas that need to be refined. You can use the alignment tool if you want to thread it yourself and 'Accept Thread' at a different point.
-
Our eighth CASP9 Refinement puzzle is the same target from 334: CASP9 Puzzle 28. Unlike other refinement targets, 2 models are suggested as starting points here. Each of these models has different bits of the structure modeled better than the other one. So that you can combine information from both of them, we are giving you both models (resetting the puzzle will randomly pick one and both are in your template reserve) and we are running the puzzle for much longer.
-
Our 37th CASP puzzle is the LAST non-refinement CASP9 target! It's a short protein with no templates so this is the final chance for Foldit to make a mark in the 'template free category' at CASP9. We are again giving you the 5 Rosetta Server predictions as starting points and templates. Try searching for different topologies rather than refining these server predictions.
-
Our 36th CASP target puzzle has 7 templates, but only 3 of them cover most of the protein. We are giving you the other 4 templates because they are reported to align well to the start of the sequence. These might be useful for partial threading.
-
Our 35th CASP target puzzle is very difficult. It was a 'server only' target until the CASP assessors noticed that the servers did very poorly on it. They have challenged us by posting: 'We want to check if human-expert groups can do better on these targets'. This puzzle has 2 templates and we are giving you 2 different alignments for each. You will find them difficult to thread, but hopefully by moving the alignments around you will be able to show that humans are better than servers!
-
Our 34th CASP target puzzle has no templates. We are giving you the 5 Rosetta Server predictions as starting points (they are also in your template reserve so that you can use partial threading). Try to look at where the different models agree and focus on the regions where they vary. You'll need to change the overall topologies, do not just refine them!
