Foldit Lab Report #12: Coronavirus Spike Binder Results

We finally have some news to report about the initial set of coronavirus Spike binders designed by Foldit players! In this video, we also share exciting progress from the researchers at the Institute for Protein Design. Watch our latest lab report now.

- Foldit Education Mode is live!
- COVID Spike binder results
- More Spike binder news


- New binder design metrics are ssllooww
- Symmetric puzzles are back in the mix

- A sweet trimer with a buried edge created by grogar7. See for yourself in the Design of the Month sandbox puzzle.

(Tue, 09/01/2020 - 16:59  |  4 comments)
Joined: 09/24/2012
Groups: Go Science
Recipes with new metrics

If I understood well, you discarded latest filters because of problems with recipes. Which problems?
I didn't experience problems. If there were problems, don't you think it's possible for cookers themselves to adapt recipes ?

For example, when filter "Move Count Limit" was introduced, it resulted in a crash of recipes using filter scores. I could solve this using the following code:

if stringfilter=="Move Count Limit" then FilterBonus=0 else
FilterBonus=filter.GetBonus(stringfilter) -- number

Suggestion: in my opinion, and in order to let Cookers use the new filters if they want for script strategies, it would be usefull to "publish" the related scores when calculated "on demand" by recipes. If the information seen on screen was available as recipe input, we'll develop strategies to use them. After all, recipes are there to avoid "repetitive tasks". Running these new metrics would be used in order to enhance our "scientific test" results (if not our score). See feedback.

bkoep's picture
User offline. Last seen 17 hours 56 min ago. Offline
Joined: 11/15/2012
Groups: Foldit Staff

Yes, we just want to make it easier for scripters to adapt to the new metrics. It seems that the new metrics were disruptive for script users, and were preventing them from playing normally.

By default, we think existing recipes should carry on as before (they should ignore the new metrics). We will be adding separate Lua functions for the long-running metrics. This way, script writers can adapt their scripts to the new metrics, and in the meantime script users can continue to play like normal.

I like your suggestion about metric scoring! We definitely want these metric values to be available for script logic.

Joined: 06/06/2013
Groups: Gargleblasters
LCB1 binder

Question for the Scientists...thanks for sharing. I'm curious that all the sidechain bond are hydrophiles. I've always tried, albeit unsuccessfully to include at least one hydrophobe on the thought that hydrophiles will bind to anything or be happy hanging out unattached. Do I need to rethink this? Is shape more of a factor? I know no one has the answers. But I'd love to get some feedback on this

bkoep's picture
User offline. Last seen 17 hours 56 min ago. Offline
Joined: 11/15/2012
Groups: Foldit Staff
Right idea

Yes I think you have the gist of it, except I disagree that polar sidechains "will bind to anything." Polar atoms must be exposed to water or make H-bonds. This applies to polar atoms on the target and the binder. If your binder creates BUNS at the interface, that is a deal-breaker and your design will not bind the target (I know the LCB1 Sandbox puzzle shows lots of BUNS, but most of these are false positives due to our BUNS approximations).

When designing a binder, I would encourage you to worry first about polar atoms at the interface. Then worry about adding hydrophobic residues to make good hydrophobic packing between the binder and target (shape complementarity is a part of this).

Also, you are right that there are lots of polar residues at the LCB1 interface, but it is not entirely polar. There are several hydrophobic MET and a couple of TYR, all of which have large hydrophobic surfaces that are buried upon binding. The burial of all this hydrophobic surface area definitely contributes to binding strength.

User login
Download links:
  Windows    OSX    Linux  
(10.12 or later)

Are you new to Foldit? Click here.

Are you a student? Click here.

Are you an educator? Click here.
Social Media

Only search
Other Games: Mozak
Recommend Foldit
Top New Users

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