Back to Recipes Homepage
recipe picture
Recipe: Rav3n_pl SBS v1.2
Created by Rav3n_pl 152 421
Your rating: None Average: 5 (8 votes)


Name: Rav3n_pl SBS v1.2
ID: 22173
Created on: Wed, 01/26/2011 - 23:26
Updated on: Wed, 06/15/2011 - 21:16

Rav3n_pl SBS Select Best Start v1 for multistart puzzle. Finding, stabilizing and fuzing up to 10 starts then loads best resoult.

Best For


Joined: 06/17/2010
Select Best Start

This script is resetting puzzle zillion times to find all starts in multistart puzzle (up to 10 of them).
Then it fuze every of found starts.
Then it loads best scoring one fater fuzes.
All save slots 1-10 are used!

Joined: 06/17/2010

+ changed fuzing algo

Joined: 11/03/2011
Groups: None
Should this work the same way

Should this work the same way if run twice against the same reset of the same puzzle?

I ran this against 519 with the following starts ...

-12947.478 -12026.923 -26263.736 -22310.880 -17981.811 -23896.052 -14984.628 -27510.257 -14284.083 -23865.093

The first run in the "main" code gave me the following results:

9711.173 9644.297 7717.018 9009.543 9714.207 9790.947 10231.703 8966.792 9098.786 9299.806

The second run gave me the following (starting at q3) ...

8570.438 9292.688 9709.803 9605.777 7829.380 (I didn't finish)

The first run in today's (Feb 21st) dev code gave me

9042.703 9611.888 8061.149 8799.215 9283.524 9640.780 9974.403 8959.807 9299.877 9064.356

The second run against the same current (2/21) developer release is also random.

Or maybe I am imagining it? None of them line up. Running the same release against the same puzzle seems to give different results.

Joined: 06/17/2010

Puzzle reset is in random order.
Also shake function can make different results each run.
In most cases script should fin same "final" score for each corresponding start position.

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



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