Placeholder image of a protein
Icon representing a puzzle

2177: Two-sided Strand Binder Design

Closed since 4 months ago

Intermediate Overall Design


July 22, 2022
Max points

Design a beta strand and a separate protein that can bind to it! This is a two-sided design problem, where we can design the sequence of both binding partners. The goal is to design a pair of proteins that will self-associate in solution. This is especially useful for bioengineering applications where we might want to use a short peptide as a "signal" that can be recognized by its binding partner. Although the short peptide may be flexible in solution, we would like to bind it in the shape of beta strand, so the target backbone has been frozen as a beta strand. However, you may still change the sidechains of the strand target to any amino acid you want!

AlphaFold predictions are enabled for the designed binder. Once you've designed a binder for the strand target, upload your solution for AlphaFold using the AlphaFold prediction tool. AlphaFold will predict the structure of your binder chain only (i.e. in the unbound state, in the absence of the strand target). If you load this prediction, then Foldit will attempt to align the prediction with your solution (i.e. in the bound state, making an interface with the strand target). If you continue working off of the AlphaFold prediction, you may need to make adjustments at the interface where the binder interacts with the target. See the puzzle comments for additional Objectives.

Top groups

  1. Avatar for Gargleblasters 100 pts. 18,648
  2. Avatar for Anthropic Dreams 2. Anthropic Dreams 60 pts. 18,452
  3. Avatar for Go Science 3. Go Science 33 pts. 18,395
  4. Avatar for Marvin's bunch 4. Marvin's bunch 17 pts. 18,078
  5. Avatar for L'Alliance Francophone 5. L'Alliance Francophone 8 pts. 17,910
  6. Avatar for Contenders 6. Contenders 4 pts. 17,908
  7. Avatar for Australia 7. Australia 2 pts. 17,162
  8. Avatar for Void Crushers 8. Void Crushers 1 pt. 16,695
  9. Avatar for SETI.Germany 9. SETI.Germany 1 pt. 15,030
  10. Avatar for Team China 10. Team China 1 pt. 9,494

  1. Avatar for stomjoh 11. stomjoh Lv 1 53 pts. 18,292
  2. Avatar for HuubR 12. HuubR Lv 1 49 pts. 18,244
  3. Avatar for Sandrix72 13. Sandrix72 Lv 1 46 pts. 18,220
  4. Avatar for phi16 14. phi16 Lv 1 43 pts. 18,215
  5. Avatar for Anfinsen_slept_here 15. Anfinsen_slept_here Lv 1 40 pts. 18,004
  6. Avatar for Punzi Baker 3 16. Punzi Baker 3 Lv 1 37 pts. 17,934
  7. Avatar for jobo0502 17. jobo0502 Lv 1 34 pts. 17,910
  8. Avatar for infjamc 18. infjamc Lv 1 32 pts. 17,908
  9. Avatar for Bletchley Park 19. Bletchley Park Lv 1 29 pts. 17,860
  10. Avatar for akaaka 20. akaaka Lv 1 27 pts. 17,855


bkoep Staff Lv 1

DDG Metric (max +2000)
The predicted binding energy of your design. The goal DDG is -40.0 or less.

Contact Surface (max +2000)
Measures how much of your binder surface is in close contact with the target strand. The goal Contact Surface is 500 or more.

Buried Unsats (max +500)
Penalizes 20 points for each polar atom that cannot make any hydrogen bonds.

Core Existence (max +1500)
Ensures that at least 20 percent of residues are buried in the core of your binder.

Interaction Energy (max +500)
Monitors that all large PHE, TYR, and TRP residues are scoring well.

SS Design (max +500)
Disabled use of CYS residues.

Neural Net (max +0)
Achieve an AlphaFold prediction confidence of 80% or more.

LociOiling Lv 1

The move tool can appear on the target strand, but it fails to do anything.

The tweak tool does not appear when the target strand is selected.

This behavior is probably the same as in other binder puzzles.

Unlike those other puzzles, it's a little difficult to see what's going on when the binder is much larger than the target.

nspc Lv 1

There is a BUNS problem. There is no BUNS on the beta strand (backbone or sidechain). Maybe we should restart the puzzle ?

nspc Lv 1

Now I have buns on the beta strand oO. Dont understand what happenned

nspc Lv 1

I added some images on the Foldit discord

strange false negatives :

BUNS comme back :

bkoep Staff Lv 1

We are not too worried about how the target strand folds. Since it is only 10 residues long, it is very likely to be unfolded and "floppy" in solution. As long as the target forms an extended sheet some of the time, then the binder will have opportunity to bind to it.

BootsMcGraw Lv 1

This is an utterly strange puzzle.

The target is a single strand, which doesn't exist in real life. Strands are found in groups of two or more. They also don't stick out from a larger protein, with both surfaces exposed to solution. What is the purpose of this exercise?

And why no ideal loops metric on this puzzle? Why do loops not matter?