Back to Recipes Homepage
recipe picture
Recipe: RaiseRigidProteinScore0
Created by jeff101 97 220
5
Your rating: None Average: 5 (2 votes)

Profile

Name: RaiseRigidProteinScore0
ID: 101689
Created on: Tue, 12/15/2015 - 15:20
Updated on: Wed, 12/30/2015 - 06:53
Description:

*0i.txt 12/29/15 1128pm code



Best For


Comments

jeff101's picture
User offline. Last seen 12 hours 39 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
What does RaiseRigidProteinScore0 (RRPS0) do?

RRPS0 freezes the protein and then uses bands to translate and rotate the protein through space as if it were a rigid body, trying always to improve the protein's overall score. In Electron Density and Symmetry Puzzles, it should be possible to raise the protein's overall score using this Recipe. Ideally, none of the protein's internal coordinates (sidechain orientations & rotamers, backbone dihedral angles, secondary structure, amino acid sequence, interatomic distances within each monomer, etc.) should change while this Recipe runs. This Recipe should help in Electron Density Puzzles when the "Center Protein on Density" button in the "Electron Density" panel does not raise the protein's overall score.

jeff101's picture
User offline. Last seen 12 hours 39 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
RRPS0 uses 2 kinds of bands:

RRPS0 uses 2 enabled (purple) bands (b1 and b2) to apply force and torque to the protein. These enabled bands' total force makes the protein translate while their total torque makes the protein rotate. Initial forces (fx1 fy1 fz1 from band b1, fx2 fy2 fz2 from band b2, and their totals fx fy fz) and initial torques (tx1 ty1 tz1 from band b1, tx2 ty2 tz2 from band b2, and their totals tx ty tz) for each loop are calculated and listed. These forces and torques use a temporary Cartesian xyz coordinate system set up by 2 atoms within the protein that are a distance r12 apart along the temporary x axis. Only forces applied perpendicular to this x axis will create torques; thus, there are never torques along the temporary x axis, so tx1=tx2=tx=0 should always hold.

RRPS0 also uses 6 disabled (gray) bands to monitor the protein motion. 3 of these bands (b1o b2o b3o) follow the motion in the latest loop while the other 3 bands (bbeg bmid bfin) follow the motion since the RRPS0 run began.

For more details about RRPS0, see its Recipe Output, its code, and the comments within its code.

jeff101's picture
User offline. Last seen 12 hours 39 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
Sample Recipe Output from Puzzle 1152:

As RRPS0 runs, it makes Recipe Output lines like below (here from Electron Density Puzzle 1152):

Total score 1,2 = 6799.5180328501, -1241.2462270735 (differ by -8040.7642599235).
Total score 1,3 = 6799.5180328501, -1241.2462270735 (differ by -8040.7642599235).
Total score 2,3 = -1241.2462270735, -1241.2462270735 (differ by 0).
Total for Clashing = -863.89436784264.
Total for Packing = 5982.0441003298.
Total for Hiding = -5208.8411917206.
Total for Bonding = 708.60130276736.
Total for Ideality = -281.18089908314.
Total for Backbone = 81.348145639469.
Total for Sidechain = -1961.1370494413.
Total for Reference = -248.16222.
Total for Density = 42.44204905652.
Total for Other = 507.53390322104.

Total score 1,2 = 7285.0532920896, -998.47859745313 (differ by -8283.5318895427).
Total score 1,3 = 7285.0532920896, -998.47859745313 (differ by -8283.5318895427).
Total score 2,3 = -998.47859745313, -998.47859745313 (differ by 0).
Total for Clashing = -863.89436784238.
Total for Packing = 5982.0441003295.
Total for Hiding = -5208.8411917205.
Total for Bonding = 708.60130276875.
Total for Ideality = -281.18089908342.
Total for Backbone = 81.348145639495.
Total for Sidechain = -1961.1370494413.
Total for Reference = -248.16222.
Total for Density = 285.20967867571.
Total for Other = 507.53390322101.

In the above, you can see that while the overall protein score "Total score 1" rose from 6799.518 to 7285.053, the "Total for Density" rose from 42.442 to 285.209, and most other scores remained virtually unchanged. That most scores remained virtually unchanged shows that in effect and as desired, the protein behaved like a rigid body while RRPS0 ran.

jeff101's picture
User offline. Last seen 12 hours 39 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
Default Input Values:

I am still studying what input values work best. I encourage you to explore input values different from the default ones and let me know what you find. Also, if you have suggestions for future versions of this Recipe, please let me know.

Thanks!

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

none

Children
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, RosettaCommons