Placeholder image of a protein
Icon representing a puzzle

2150: CD47 Binder Design: Round 9

Closed since 4 months ago

Intermediate Overall Design


May 20, 2022
Max points

Design a protein that can bind to CD47! This puzzle has a reduced BUNS penalty so that players can focus on other Objectives--but don't forget about BUNS completely! AlphaFold predictions are available for your designed binder. Once you've designed a binder for the 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 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 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.

CD47 is a protein displayed on surface of human cells. Normally, CD47 is presented to the immune system as a "friendly" signal, and can prevent an over-zealous immune system from attacking healthy human tissue. However, some types of cancer will over-produce CD47 as a way to evade the immune system, camouflaging a tumor as "friendly" human tissue. A protein that binds to CD47 might be useful for treating these types of cancer. We'd like to design a protein binder that can block CD47 recognition by the immune system, and allow immune cells to attack tumors in the body.

In this puzzle, we've presented a section of the CD47 target that is recognized by the immune system. The backbone and most of the sidechains are frozen, except at the binding site, where flexible sidechains are shown in color. Players can design a new protein that can bind to these sidechains and block CD47 recognition. In order to bind the target, designs will need to make lots of hydrophobic contacts and satisfy any polar atoms that are buried at the interface. But designs will also need to have lots of secondary structure (helices or sheets) and a large core, so that they fold up correctly! See the puzzle comments for Objective details.

Top groups

  1. Avatar for Anthropic Dreams 100 pts. 14,297
  2. Avatar for Go Science 2. Go Science 56 pts. 14,068
  3. Avatar for Marvin's bunch 3. Marvin's bunch 29 pts. 13,897
  4. Avatar for Contenders 4. Contenders 14 pts. 13,694
  5. Avatar for Gargleblasters 5. Gargleblasters 6 pts. 12,785
  6. Avatar for Australia 6. Australia 2 pts. 12,392
  7. Avatar for L'Alliance Francophone 7. L'Alliance Francophone 1 pt. 11,567
  8. Avatar for Beta Folders 8. Beta Folders 1 pt. 8,960
  9. Avatar for Team China 9. Team China 1 pt. 5,079
  10. Avatar for Foldit Staff 10. Foldit Staff 1 pt. 0

  1. Avatar for LociOiling
    1. LociOiling Lv 1
    100 pts. 14,288
  2. Avatar for ichwilldiesennamen 2. ichwilldiesennamen Lv 1 94 pts. 14,060
  3. Avatar for gmn 3. gmn Lv 1 89 pts. 13,925
  4. Avatar for dcrwheeler 4. dcrwheeler Lv 1 83 pts. 13,908
  5. Avatar for fpc 5. fpc Lv 1 78 pts. 13,895
  6. Avatar for Sandrix72 6. Sandrix72 Lv 1 73 pts. 13,724
  7. Avatar for alcor29 7. alcor29 Lv 1 69 pts. 13,715
  8. Avatar for phi16 8. phi16 Lv 1 64 pts. 13,662
  9. Avatar for nspc 9. nspc Lv 1 60 pts. 13,655
  10. Avatar for Bruno Kestemont 10. Bruno Kestemont Lv 1 56 pts. 13,633


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 protein. 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 25 percent of residues are buried in the core of the monomer unit.

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

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)
Disabled use of CYS residues.

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