1613: Symmetric Dimer Design: Hydrogen Bond Networks

Closed since about 7 years ago

Summary

• Created: December 21, 2018
• Expires: December 28, 2018 at 23:00 UTC
• Max points: 100

Description

This symmetric design puzzle has C2 symmetry, with two symmetric chains. The H-bond Network Objective encourages players to bury H-bond Networks at the interface between the two chains. We've doubled the H-bond Network bonus, but we'd like players to focus on building networks deep in the core of the protein complex. There are a couple other bonuses in effect; see the puzzle comments for details. The Baker Lab will run folding predictions on your solutions for this puzzle, and those that perform well will be synthesized in the lab. Remember, you can use the Upload for Scientists button for up to 5 designs that you want us to look at, even if they are not the best-scoring solutions!

Top groups

• 1. Contenders 100 pts. 26,979
• 2. Beta Folders 78 pts. 26,827
• 3. Anthropic Dreams 60 pts. 26,150
• 4. Gargleblasters 45 pts. 25,958
• 5. Go Science 33 pts. 25,910
• 6. L'Alliance Francophone 24 pts. 25,806
• 7. Void Crushers 17 pts. 25,058
• 8. Marvin's bunch 12 pts. 24,613
• 9. Russian team 8 pts. 23,386
• 10. FoldIt@Netherlands 6 pts. 23,099

Top soloists

• 1. Mark- 100 pts. 26,932
• 2. LociOiling 97 pts. 26,652
• 3. Wojcimierz 93 pts. 26,467
• 4. Glen B 90 pts. 26,436
• 5. retiredmichael 86 pts. 26,198
• 6. Galaxie 83 pts. 26,148
• 7. pvc78 80 pts. 26,008
• 8. actiasluna 76 pts. 25,958
• 9. johnmitch 73 pts. 25,957
• 10. grogar7 71 pts. 25,930

Comments

[bkoep]

Core Existence: Monomer (max +2800)
Ensures that at least 28 residues are buried in the core of the monomer unit. The threshold of "buriedness" between core and surface residues is more stringent than usual, so residues may have to be "extra buried" to be counted toward the core.

Core Existence: Complex (max +3400)
Ensures that at least 68 residues are buried in the core of the entire complex (including the interface between monomer units).

HBond Network
Rewards networks that comprise at least three hydrogen bonds involving core residues. Networks must be at least 70% satisfied (i.e. 70% of all oxygens and polar hydrogens in a network must make a hydrogen bond).

SS Design (max +500)
Prohibits all CYS residues. Prohibits GLY, ALA residues in sheets and helices.

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.

[spvincent]

It seems there's a recent paper from the Baker group on this subject summarized here:

ScienceDaily

Scientists program proteins to pair exactly

Proteins designed in the lab can now zip together in much the same way that DNA molecules zip up to form a double helix. The technique could enable the design of protein nanomachines that can potentially help diagnose and treat disease, allow for the...

It looks like Rosetta can now design these hydrogen bond networks to a high level of accuracy. Which is great of course but does leave me wondering what they're hoping to learn from this current crop of dimer puzzles

[bkoep]

I'm sorry I didn't get to around to answering this before the puzzle closed!

The Rosetta tool for finding H-bond networks is very good at mutating residues to design a network onto a fixed backbone. The typical Rosetta workflow is to generate a library of protein backbones, and then look for H-bond networks in each backbone. Most backbones cannot accommodate a fully-satisfied network—but that's okay, as long as we have a large number of backbones to scan through.

We think that Foldit players might be able to find H-bond networks with modifications to the protein backbone. If players happen upon a backbone that supports an almost-satisfied network, then players might be able to see how the backbone could be adjusted to complete the network. Rosetta isn't very good about imagining what could be (it only sees the exact backbones that are fed to it), so this could be a complementary approach to the design of H-bond networks.