Back to Recipes Homepage
recipe picture
Recipe: nicm25.dei.TestBaseScoring V0
Created by Nicm25 70 133
0
Your rating: None

Profile

Name: nicm25.dei.TestBaseScoring V0
ID: 105164
Created on: Mon, 08/30/2021 - 05:14
Updated on: Mon, 08/30/2021 - 12:14
Description:

*demo script, test base scoring! (segment select, number seed, number setcount, boolean fixrandom)



Best For


Comments

Nicm25's picture
User offline. Last seen 10 hours 37 min ago. Offline
Joined: 04/22/2020
Groups: None
TestBaseScoring description

*note: this is demo script.
This recipe just repeats the random mutation and quick relax to find out the base energy required for folding,
and recipe has been adjusted to run in the 'Protein Design Sandbox'.
before running you should give secondary structure to you protein, examine it using idealSS as initial structure by this recipe.
4 arguments
select: currently selected segment will be this argument.(check mode only)
seed: choose any value you like to generate random AA's by random seed.
range: 0.0-1.0, default: 0.5, function: 1.0 is plus value, 0.0 is none value
setcount: run 20 times as one set, only 0 is check mode for scoring only.
range: 0-50 (ext0-1000), default: 1, function: 1 is one set 20 counts, 0 is check mode
fixrandom: set whether to fixed random seed.
default: false, function: false is always random seed, true is fixed random seed

Nicm25's picture
User offline. Last seen 10 hours 37 min ago. Offline
Joined: 04/22/2020
Groups: None
Documentation

Hi, this is an explanation of how this recipe was created....

I was trying to investigate in this recipe was true(valid) energy required for folding.
If we could pick arbitrary AA's and have our folding trends known, we would be able to design more stable energy landscape.
It is known that we have tendency to spontaneously fold from single strand(beta) into helix(alpha) where hydrogen bonds exist.
In order fold to helix, we must find another energy landscape that is more stable from helix.
to bend the helix, we have to find another energy landscape that makes helix more stable,
such as AA's sequence that is more stable when attached to each other's side chains by hydrophobic interactions.
How do we do this? using that result table, we can deduce from delta score the energy required to go from helix to strand.
and to bend the helix, we as loop 2-3 segments, some of which are untied from helix, and re fold them as strand.
this means that we need hydrophobic sidechain that gives high score in the regular case, least about 100 points(10 kcal/mol).
and when conformation can be altered by surface hydrogen bonding etc..., it may be re folded for more stable, lower energy.
*however, since that result table does not include hydrogen bonds, there is offset of about 10 points per segment in delta score.

*attached data
the results of helix and strand(sheet) data for each sidechains.

results tabulated using this recipe are as follows.

AAIndex,samples,helixscore,sheetscore,deltascore
A,3758,10.262,-61.200,71.462
C,3759,-61.791,-107.314,45.523
D,3832,24.937,3.691,21.246
E,3782,27.563,10.892,16.671
F,3754,-21.907,-18.889,-3.018
G,3759,5.880,-48.695,54.575
H,3865,14.048,-0.313,14.361
I,3683,-25.382,-50.216,24.834
K,3776,26.419,8.902,17.517
L,3748,-21.663,-63.682,42.019
M,3762,-23.315,-50.707,27.392
N,3772,22.178,-1.865,24.043
P,376,4.343,42.710,-38.367
Q,3746,21.790,2.002,19.788
R,3714,17.443,-2.977,20.420
S,3697,21.226,-16.516,37.742
T,3805,20.256,-35.937,56.193
V,3856,-22.929,-74.463,51.534
W,3824,-31.079,-46.403,15.324
Y,3732,-10.375,-2.023,-8.352
*sigma=0.18
Nicm25's picture
User offline. Last seen 10 hours 37 min ago. Offline
Joined: 04/22/2020
Groups: None
this recipe is outdated.

New recipes are available here
https://fold.it/portal/recipe/105209

that difference is added arguments, and testing score in shape at start of this recipe using Quick save and load(1).

Want to try?
Add to Cookbook!
To download recipes to your cookbook, you need to have the game client running.
Parent

none

Children

none

Authors
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