DEVPREV] Loop Remodeling

Case number:699969-2004325
Opened by:Mike Cassidy
Type:DevPrev Issue
Opened on:Saturday, October 14, 2017 - 00:20
Last modified:Monday, October 16, 2017 - 23:01

I have frozen twice. Also I can not move or so do anything to the ligrand; is that correct?

(Sat, 10/14/2017 - 00:20  |  7 comments)

Joined: 06/24/2008
Groups: Void Crushers

Now froze three times.

Susume's picture
User offline. Last seen 1 hour 38 min ago. Offline
Joined: 10/02/2011

I'm not sure I see the point of having all the cutpoints if we can't make the loops shorter or longer. Add and delete segments are not allowed.

Joined: 01/15/2010
Groups: None

The cutpoints are there such that backbone changes in the loops don't propagate to the fixed portion of the protein. (Without them a global wiggle could result in the fixed portion of the protein falling apart.)

Susume's picture
User offline. Last seen 1 hour 38 min ago. Offline
Joined: 10/02/2011

I'm no chemist, but the bond counts in the ligand look weird to me. The 2 hexagons and one pentagon are drawn with all double bonds (in any of the cartoon or stick views), with the oxygen in the hexagon having 4 bonds and the carbons having 6 or 5. In the pentagon that is in a different plane from the hexagons, two of the carbons have only 3 bonds. The oxygen that is bonded to residue 131 seems to have a carbon hanging off it with no hydrogens, so that carbon appears to have only 1 bond - unless that is supposed to be an OH, in which case the H did not get painted white (and is strangely long).

I don't know if any of this affects playability - it just bothers me.

Joined: 01/15/2010
Groups: None

As you might have guessed by now, this devprev puzzle was a preview of 1440: Aflatoxin Challenge: Round 1. The ligand is supposed to be aflatoxin B1.

You're correct that the display of the rings is off. The 5-member ring with the two carbons with just one hydrogen each should indeed have a double bond between those two carbons. The other rings have all double bonds because they're annotated as "aromatic", and Foldit represents aromatic bonds as double bonds. (Note that the 5-member ring on the other side, the one with the two carbons each with two oxygens, actually shouldn't be aromatic.)

None of this affects the puzzle play, though. The single/double/aromatic bond annotation here is purely for display. Everything that affects the scientific results (and puzzle scoring) is annotated correctly, as far as I can tell. (But you're right that it can be distracting.)

The oxygen bonded to 131 is a bit of an interesting case. That's actually not a carbon or hydrogen that's bound to it with a strangely long bond, but a calcium ion. This is part of the active site of the enzyme. The ligand being modeled isn't actually aflatoxin B1 proper, but a transition state model of the aflatoxin degradation reaction. The calcium is interacting with that oxygen and helping to stabilize the transition state for a reaction where water is breaking apart the aflatoxin molecule. So that helps explain the "strange" geometry there. (It's more a coordination bond than a single bond, but Foldit has limitations on how it can display bonds.)

Joined: 09/24/2012
Groups: Go Science

Everything ok with recipes that include very diverse functions.

Joined: 01/15/2010
Groups: None

Not being able to move the ligand is deliberate. The hope is to design a protein which will bind the ligand in that particular geometry with respect to certain residues in the fixed portion of the protein.


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