h-bond network scoring

Case number:699969-2004942
Topic:General
Opened by:fiendish_ghoul
Status:Open
Type:Bug
Opened on:Tuesday, March 27, 2018 - 22:18
Last modified:Friday, March 30, 2018 - 23:46

H-bond network has only two bonds but is still credited (according to puzzle comments it must have at least three bonds).

Shared solution with scientists.

(Tue, 03/27/2018 - 22:18  |  2 comments)


bkoep's picture
User offline. Last seen 23 min 40 sec ago. Offline
Joined: 11/15/2012
Groups: None

It looks like the hydrogen of the tyrosine is pointing at the hydrogen of one of the serines, so that the serine is "bonding" to the tyrosine as both an acceptor and donor to count for two bonds. This should not score so well; I'll dig into it a little more. Thanks for pointing it out!

bkoep's picture
User offline. Last seen 23 min 40 sec ago. Offline
Joined: 11/15/2012
Groups: None

Just to follow up, apparently this is a known issue in Rosetta, but happens rarely enough that it's usually not a problem.

To be clear, this is physically unrealistic—a hydrogen bond could never form with two hydrogens pointing at each other like this. But when a hydroxyl serves as an H-bond acceptor, Rosetta does not consider the position of its hydrogen (that is, the hydrogen that is not participating in the H-bond).

There is a fix for this in development, but it will be some time before that makes it into Foldit. In the meantime, we'd appreciate if you can keep an eye out for this problem, and share with us whenever you encounter it! If Foldit players encounter this problem frequently, we want to know about it!

Sitemap

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