Residue Count (max +275)
Penalizes extra residues inserted beyond the starting 152, at a cost of 55 points per residue. Players may use up to 157 residues in total.
Core Existence (max +1000)
Ensures that at least 25 percent of residues are buried in the core of the monomer unit.
Ideal Loops (max +500)
Penalizes any loop region that does not match one of the Building Blocks in the Blueprint tool. Use "Auto Structures" to see which regions of your protein count as loops.
SS Design (max +500)
Penalizes all CYS residues. Penalizes GLY, ALA residues in sheets. Penalizes GLY, ALA, SER, THR in helices.
Thank you, bkoep, for (again) explaining the objectives that can give us bonus points. And thank you, and all of your fellow scientist working on this project, for your outstanding work!
Starting the puzzle with quite a high score is an interesting concept. I assume the reason why the initial score is so high, is because this protein was designed and folded by a very renowned player (see the puzzle description). Now the challenge to all of us is to improve that score.
In the (few) puzzles I played this far it was often not difficult to get the score well above the starting value by using a few simple actions, like apply and idealize secondary structure, shake, wiggle, but now all of that has already been done. I hope this does not scare off too many players, because it seems to be very hard to "get off the mark".
Starting at this high score might be a good way to encourage players to try different concepts for the secondary structure. In fact, that is what I did initially. I have a habit (in puzzles as well as in real life) of not always going with the crowd. Although I acknowledge that teamwork can be very powerful, I sometimes feel that there are already enough pairs of eyes looking at something from the same viewpoint, and then I tend to try a different perspective. There was an interesting discussion in global chat yesterday about teamwork, and about working in a local optimum versus trying to find the global optimum. I think it was LociOilingIRC who raised that point.
Having said all that, can I ask you, bkoep, or one of the other scientists, a question?
I wonder whether the score that we get in this puzzle is a good indication of how well a certain protein will stick to the binding site. Correct me if I'm wrong, but isn't most of this score made up of how well the protein itself was built up and folded?
What I tried is the following. Starting from the initial design (reset the puzzle), I moved the protein away from the binding site, and then tried to find a new (local) optimum by shaking and wiggling. In my mind, the number of points that you lose this way (or in other words, the number of points gained by placing the protein at the binding site and re-optimizing it there) would be an indication of how "sticky" the protein is. Again, please correct me if I'm wrong. I am not a molecular biologist, just an electrical engineer who likes puzzles.
Could you give us an idea why the score (and bonus points) were set up the way they are? Thanks!
Looks like you had already answered my question, long before I asked it, in your blog post Analysis of protein binder designs. Thanks! Lots of interesting stuff in there, way more than I could ask for.
This gives me enough incentive to stop focussing on the score and go my own way. Don't know whether it will get us anywhere, but the least I can do is try.
Now there's only one question left: if I would indeed find a "sticky" protein, but with a Foldit score that is lower than most other players', how can I make sure that this design is noticed by you, the scientists? Should I just save a solution and "Upload For Scientists" (in the Open/Share Solutions dialog box)?
Yes! That's exactly what the Upload for Scientists button is for. If you think your solution is promising, but it doesn't make the leaderboards, use Upload for Scientists and we'll be sure to take a look.
Great questions, glad you were able to find our previous blog post!
Not sure if that was a question, but that's the secondary structure for each segment, using the codes:
The recipe print protein and several others use this format.
Actual science tools like JPred use a similar format, but with "-" instead of "L" to indicate loop.