puzzle picture
271: Hold it! 2
Status: Closed

Summary

Name: 271: Hold it! 2
Status: Closed
Created: 03/09/2010
Points: 100
Expired: 03/15/2010 - 23:00
Difficulty: Advanced
Description: Excellent work on the previous Hold it! puzzle! After austinday analyzed your solutions, we've made a few changes to the setup for this new round. In particular you should try to: 1) avoid mutations to Arginine, Lysine, Aspartate, and Glutamate. 2) make at most three hydrogen bonds at the very tip of the ligand. 3) make even more ideal hydrogen bonds. When they are unideal, you'll notice the unideal angles will appear as red lines. The green line near them shows a more ideal angle! We're starting you off with some interesting designs that came out of the previous round so you don't have to start from scratch. Again, we don't know if it's possible!
Categories:

Top Groups

Top Evolvers

Top Soloists



Need this puzzle? Log in to download.  

Comments

keypad5's picture
User offline. Last seen 5 years 36 weeks ago. Offline
Joined: 11/09/2009
Mutations and scoring

A couple of questions:

Firstly, how strictly do we need to avoid those listed sidechains? So far I haven't been able to find a solution that doesn't use at least one of those listed sidechains to bond to the end of the ligand. (but that doesn't mean I'll stop trying to find a solution!!)

Secondly, it seems likely that a higher score can be achieved by using some of those sidechains that are to be avoided. e.g., I can run a global mutate to increase my score but I know that it'll probably change a number of my sidechains to the ones that ought to be avoided. If I have a solution that doesn't score as well as my best scoring solution, but fulfils the requirements, how do I submit it?

K5

keypad5's picture
User offline. Last seen 5 years 36 weeks ago. Offline
Joined: 11/09/2009
Follow up

Ooh, I just realised I'm using asparagine, not aspartate. :-) I've been trying to avoid glutamine and asparagine by mistake.

Still, I think it's a valid concern that a global mutate will raise the score but not abide by the requirements, which means our 'best' or 'most appropriate' solutions may not be our best scores.

Joined: 11/10/2007
Groups: Window Group
You are correct. However,

You are correct. However, many of the automatic algorithms only know about the score, and don't take the conditions into account. It is true that there may better scoring solutions that don't meet the conditions. I think there has been a request for "restore best with conditions met", would that be useful?

keypad5's picture
User offline. Last seen 5 years 36 weeks ago. Offline
Joined: 11/09/2009
That could be helpful,

That could be helpful, although it is still possible to meet the conditions of this puzzle while using the 'illegal' residues (I think), so even restoring to "best with conditions met" could still restore the puzzle to a solution with an arginine attached to the ligand end. Could it not?

I chatted with austinday in game and he said that you guys check over all of our progress, not just our final, highest-scoring solutions. If that's the case then I think my concerns are covered because you'll be able to find our potentially lower scoring solutions that have better residue combinations.

Brick's picture
User offline. Last seen 19 weeks 3 days ago. Offline
Joined: 07/15/2008
Groups: Beta Folders
Does anyone find it at all

Does anyone find it at all disturbing that only 3 players are even able to find a solution that meets the requirements to even get a score?

Why bother working on it - if you're number 4 you get one point?

infjamc's picture
User offline. Last seen 3 years 1 day ago. Offline
Joined: 02/20/2009
Groups: Contenders
Are you sure that you'll even get one point?

Brick: wait a minute...

Per Madde's comment at http://fold.it/portal/node/987447 , I'm afraid that we're talking about zero points here (so far, only 5 scores registered, and supposedly you need 10 for a puzzle to score at all)

beta_helix's picture
User offline. Last seen 1 day 2 hours ago. Offline
Joined: 05/09/2008
Groups: None
we will give you points for both Hold It! puzzles

This "zero point" problem is indeed a bug (I just replied to the feedback tracker about this, http://fold.it/portal/node/987447#comment-6669) and we will assign credit for these 2 puzzles retroactively.

We apologize for that, and we realize how difficult these 2 biotin design puzzles have been.
We are trying to think of ways to make them more fun, and not quite so hard, but it is a difficult problem that we are trying to solve.

In terms of points, though, we will make sure everyone who played the Hold It! puzzles (and groups) get credit for it, but it might take us a while so please be patient.

Thank you!

steveB's picture
User offline. Last seen 3 years 6 weeks ago. Offline
Joined: 01/08/2009
Groups: Void Crushers
Suggestions

To make this puzzle more playable, yet still yield results could I suggest the following :

1) Increase the allowable angle tolerances on the bonds. I guess they are set at about 1 degree at present...increase them to about 4 or 5 degrees
2) Score the puzzle with a bonus for the accuracy of the bonds. For example score 9500 for the structure plus
1 bond within 1 degree = +1000 pts
1 bond within 2 degrees = +300 pts
etc

That way you will get a lot more people on the scoreboard and we can still refine the bond angles using the green and red traingle thingy.

Joined: 11/10/2007
Groups: Window Group
Yes, steveB, we are planning

Yes, steveB, we are planning on implementing a bonus system similar to what you suggest. Also, I believe the angle tolerances on this puzzle are +/- 25 to 35 degrees depending on the particular angle.

Joined: 05/03/2009
Groups: Contenders
server's full

...so here's the pic

steveB's picture
User offline. Last seen 3 years 6 weeks ago. Offline
Joined: 01/08/2009
Groups: Void Crushers
Another suggestion

One of the difficulties in playing this puzzle is that as you get the bond close to the desired angle, the programme automatically switches the alignment triangle over to another bond which is further out of line than the one you are working on. This, amongst other things, drives you mad.

Would it be possible to implement in the 'conditions' section a seperate line for each bond, along with a 'show' check box for each bond...and have the traingle alignment tool remain with the selected bond ?

Get Started: Download
  Windows    OSX    Linux  
Windows
(7/8/10)
OSX
(10.7 or later)
Linux
(64-bit)

Are you new to Foldit? Click here.

Are you a student? Click here.

Are you an educator? Click here.
Search
Only search fold.it
Recommend Foldit
User login
Soloists
Evolvers
Groups
Topics
Top New Users
Sitemap

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, RosettaCommons