sss full of glycine,

Case number:699969-2012868
Topic:General
Opened by:alcor29
Status:Open
Type:Bug
Opened on:Tuesday, March 8, 2022 - 01:04
Last modified:Thursday, March 10, 2022 - 00:40

When I run mnw and disallow glycine and alanine it still mutates to helices and sheets full of glycine and perhaps too much alanine also. I remove it and it comes back.

(Tue, 03/08/2022 - 01:04  |  6 comments)


alcor29's picture
User offline. Last seen 16 hours 11 min ago. Offline
Joined: 11/16/2012

For example I have a high scoring fold in which a helix of 15 aas, 10 are alanine. I have tried about six scripts and more to try to get a better mutate but mutate insists that this is the best helix. AF of entire fold is 63%

bkoep's picture
User offline. Last seen 18 hours 9 min ago. Offline
Joined: 11/15/2012
Groups: Foldit Staff

I'm not sure this is a bug. Are you able to find manual mutations that score higher than Foldit Mutate?

You may need to adjust the protein backbone. If the backbone is packed too closely to other parts of the protein, then some mutations will introduce clashing. ALA and GLY are the two amino acids with the smallest side chains; it could be that these are the only amino acids that do not cause clashing.

If this is the case, you could also try setting Clashing Importance to zero before using Mutate. This will definitely create new clashes, but might yield more "helix-like" mutations.

alcor29's picture
User offline. Last seen 16 hours 11 min ago. Offline
Joined: 11/16/2012

Did some experimenting. Both from the current aa setup and by setting all aas to serine prior to mutating using hotkey M. CI 0.00 loads ninety percent as tryptophan with either method.Tried CIs from 0.9 to 0.1 at 0.1 intervals and still got a lot of alanine/glycine.

Wont bore you with the rest. I do have a question. Can you tell me which aa has the median sphere? That way I can load it before designing in an effort to maintain a better spacing between sss.

alcor29's picture
User offline. Last seen 16 hours 11 min ago. Offline
Joined: 11/16/2012

Also, another question. Is it accurate to say then that the MNW script does not override the foldit allowed aa load?
I have always felt that the script followed the selections I checked. But, maybe it just seemed that way because the foldit load in those days did not allow alanine and glycine in the sss?

alcor29's picture
User offline. Last seen 16 hours 11 min ago. Offline
Joined: 11/16/2012

And, I did a little checking and in order to get less As and Gs by moving the SSs as suggested, I have to loosen up the core and give up core points.

alcor29's picture
User offline. Last seen 16 hours 11 min ago. Offline
Joined: 11/16/2012

Am aware that we always have to make trade offs since we often are in a whack-a-mole situation but maybe the core requirement could use a little tweaking?

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