CIS peptide bond fixer too aggresive

Case number:954892-995590
Topic:Developer Preview
Opened by:AsDawnBreaks
Status:Closed
Type:DevPrev Issue
Opened on:Friday, July 19, 2013 - 14:06
Last modified:Tuesday, December 29, 2020 - 12:01

I click on the button to fix, and it doesn't just tweak the area a bit, it swings my whole protein around. Rebuild works better to fix these now, as it only moves the area by a small amount. See screenshots (fixing just one issue).

AttachmentSize
Before CIS.png251.93 KB
After CIS.png252.33 KB
(Fri, 07/19/2013 - 14:06  |  12 comments)


brow42's picture
User offline. Last seen 1 year 23 weeks ago. Offline
Joined: 09/19/2011
Groups: None

I would suggest making a cut at the next u-turn or proline or glycine, probably downstream (but maybe upstream?). The 180 degree twist has to go somewhere.

Joined: 04/15/2012
Groups: Beta Folders

Just for fun, thought I'd mention using the new tool to fix all gave me a nearly straight protein. Rebuild works fine as an alternative.

Angus's picture
User offline. Last seen 5 hours 25 min ago. Offline
Joined: 06/04/2008
Groups: Beta Folders

This new CIS "fix" tool doesn't seem to be good for anything, just flips half the protein somewhere off to never-never land.

Joined: 04/15/2012
Groups: Beta Folders

Good as an indicator though. It should just move locally instead.

brow42's picture
User offline. Last seen 1 year 23 weeks ago. Offline
Joined: 09/19/2011
Groups: None

Indeed, at least now you know it's there...rebuild will eventually eliminate these (by creating a twist) if you don't want Foldit to fix it for you. Before you'd never know they were there unless you were really really skilled at atom view (itskimo was godly at that).

I made a banding script to encourage creation/elimination of these bonds, but rebuild rejects so many poses when atoms are banded that I abandoned that approach.

Joined: 09/21/2011

i band key locations or run zlb, and freeze my structures, then when it swings out just shake wiggle to restore protein.

beta_helix's picture
User offline. Last seen 2 hours 15 min ago. Offline
Joined: 05/09/2008
Groups: None

A little more info about the tool in the comments of the blogpost:
http://fold.it/portal/node/995580#comment-24061

jflat06's picture
User offline. Last seen 4 hours 10 min ago. Offline
Joined: 09/29/2010
Groups: Window Group

Any time you change a backbone degree of freedom, it necessarily moves every residue down the chain. We get around this with a lot of the Foldit tools by inserting cuts and then resolving them behind the scenes. We already have tools for doing this, such as rebuild (as other people have mentioned).

This tool was introduced as a means to fix the problem directly, while letting the player deal with the side effects. The tool can't really be aggressive - there are 2 values for the omega angle in a peptide bond. Cis and trans. There is not in-between.

Brow's suggestion of inserting a cut somewhere down the chain is a good strategy for mitigating the side effects of the tool.

Joined: 04/15/2012
Groups: Beta Folders

Ok, I see. But could we have something that changes just enough to resolve it, not make it the opposite? Or even just make it an indicator you can turn on and off (like clash). It does not good to use if the protein is destroyed.

jflat06's picture
User offline. Last seen 4 hours 10 min ago. Offline
Joined: 09/29/2010
Groups: Window Group

The problem is that the only valid values for the omega angle are 180 degrees (trans) and 0 degrees (cis). The values between them aren't really valid, so there is no fixing it 'just enough'.

The primary intent of the visualization was just that - a visualization. You can always choose to ignore the click-to-switch feature and resolve the issue in your own way, such as with a rebuild. You can turn off the indicator by checking the "Show backbone issues" checkbox in the visualization options menu.

Joined: 07/21/2013
Groups: Beta Folders

I think that the problem is that the cis isomers are created by the wiggle tool. Fix wiggle so it does not create the cis isomer. Proline is an exception to this it can have cis isomers.

Joined: 09/24/2012
Groups: Go Science
Status: Open » Closed

outdated

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, Boehringer Ingelheim, RosettaCommons