1054: Vancomycin Binder: Shape Complementarity

Closed since about 11 years ago

Summary

• Created: February 13, 2015
• Expires: February 27, 2015 at 06:00 UTC
• Max points: 100

Description

Design a binding pocket for vancomycin! Antibiotic resistance is an increasing problem and one of the last resorts for treating inflammation is the cyclic peptide vancomycin. Hence there is a general interest to generate proteins that can bind vancomycin and upon binding generate a signal. With this technology it would be possible to test patients that have been treated with antibiotics if they still have small amounts left before they leave e.g. a hospital. The aim of this puzzle is to optimize the interaction between a protein and the small molecule vancomycin.

This puzzle also introduces the new Shape Complementarity filter. Shape complementarity is a measure of how well the ligand's surface contours match the cavity of the binding pocket, and has been shown to be a good discriminator between binders and non-binders. Use the "Show Isosurface" view option to visualize the protein surface, and increase shape complementarity to improve your score!

Top groups

• 11. Deleted group pts. 12,751
• 12. SETI.Germany 6 pts. 12,731
• 13. Russian team 4 pts. 12,727
• 14. Italiani Al Lavoro 3 pts. 12,603
• 15. FoldIt@Netherlands 2 pts. 12,536
• 16. Natural Abilities 2 pts. 12,520
• 17. It's over 9000! 1 pt. 12,491
• 18. xkcd 1 pt. 12,436
• 19. TH2014 1 pt. 12,310
• 20. freefolder 1 pt. 12,233

Top soloists

• 1. BitSpawn 100 pts. 12,896
• 2. retiredmichael 99 pts. 12,888
• 3. cobaltteal 97 pts. 12,871
• 4. gmn 95 pts. 12,870
• 5. mirp 93 pts. 12,869
• 6. Deleted player pts. 12,868
• 7. christioanchauvin 90 pts. 12,864
• 8. Bletchley Park 88 pts. 12,855
• 9. johnmitch 86 pts. 12,851
• 10. dcrwheeler 85 pts. 12,845

Comments

[BitSpawn]

and cut works too :-?

[Skippysk8s]

I have not found the cloud. That might be good. I would like a tool that lets me see sidechain links between the ligand and main protein only, not all sidechain links. If we have a lot of these puzzles, it would be very helpful. this is running slowly and the display is very jerky due to the large puzzle size…. simpler info of the right kind would be useful

[jamiexq]

If the purpose is to focus on and change (rebuild) the pocket - then why can't we idealize those segments?

[pauldunn]

I can't find "Shape Complementarity" in the return value for puzzle.GetPuzzleSubscoreNames. Is this intentional?

[viosca]

I noticed that it does appear in the notes.

[stomjoh]

I know the comment says to look at other orientations, but if the ligand is completely outside of the pocket adjacent to the perimeter/arc of the protein I wouldn't expect to see a Complementarity score of +1459.2. Is that really right?!?

[retiredmichael]

not sure if this is going to work, team pattern is consistent, highest scores are outside the pocket,

after getting a good position with some side-chain interaction, shake side-chains and wiggle causes it to move away from the pocket,

tried running an idealized script, mutate no wiggle, c&ws early in order to stabilize the protean. it was better and will share to scientists if you like, but would not stay in contact.

[spmm]

Is turned on and off in the view menu, it is more of a translucent skin covering the bones to show the surface than an unattached cloud. We cant see hydrophobic bonds anywhere - not just about hydrogen bonds

[Skippysk8s]

I got good bonds down one side of protein (5, with both ends and 3 in middle). banded them. will share, but best score is to abandon the bonds. I have a smaller machine and team. I'm sure some bonds in the middle of the other side too might have helped. but this puzzle doesn't work as intended. any more like it seems like a waste of our time. perhaps scientists can post here after the close.
obviously 5 bonds spread out are not enough to hold over the clashing penalty. what is? need to rethink this.