Back to Recipes Homepage
recipe picture
Recipe: Tvdl enhanced DRW 1.2.4b
Created by jeff101 97 220
4.416665
Your rating: None Average: 4.4 (24 votes)

Profile

Name: Tvdl enhanced DRW 1.2.4b
ID: 45032
Created on: Thu, 12/20/2012 - 13:57
Updated on: Mon, 12/24/2012 - 04:15
Description:

Ravs DRW with a lot more features. Selections, mutate options and a lot faster. Read the comment for more info. This version the starting clash importance is a factor for ALL clashimportance dependend parts. Fixed a score evaluation bug on slot 5 and qStab will allways return the best score during it. Included a warning if running on a low CI. Fixed missing a high score during rebuild phase.



Best For


Comments

jeff101's picture
User offline. Last seen 3 hours 3 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
New features in Tvdl enhanced DRW 1.2.4b:

This was adapted from Tvdl enhanced DRW 1.2.4 (http://fold.it/portal/recipe/43754). Below are its new features:

(A) At regular intervals it now outputs the time, the protein's sequence and secondary structure, and, if there are bands in use, how far these bands are from their desired lengths. Listing the protein's sequence should help if DRW makes mutations while it runs.

(B) It lists the scores used to determine which regions get rebuilt next.

(C) It now has 3 methods for determining the order that regions get rebuilt:

Method 0 is like in previous versions of DRW and is the default. It rebuilds the worst-scoring regions first.

Methods 1 and 2 generally rebuild the worst-scoring regions first, but they postpone doing overlapping regions and try to sample more parts of the protein sooner, a bit like playing the game Battleship or looking for a needle in a haystack. The region that will gain the most points isn't always the worst-scoring region.

A slider to choose between Methods 0-2 is in the "More options" menu.

jeff101's picture
User offline. Last seen 3 hours 3 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
Method 1: shown here for len 3 on a protein with 14 AA's:

Make 3 staggered sets of groups with 3 consecutive AA's in each group (below lists the AA or segment #'s in each group):

Set A: 1-3 4-6 7-9 10-12
Set B: 2-4 5-7 8-10 11-13
Set C: 3-5 6-8 9-11 12-14

Find the rank order that each of the 12 groups above would usually be done
(rank 1 would usually be done 1st, rank 12 would usually be done last),
for example:

Set A: Ranks for each group: 1 12 3 6 Average Rank = 22/4
Set B: Ranks for each group: 8 2 9 7 Average Rank = 26/4
Set C: Ranks for each group: 4 10 11 5 Average Rank = 30/4

Set A has the lowest average rank, so its groups all get done 1st.
Set B has the next lowest rank, so its groups all get done 2nd.
Set C has the highest average rank, so its groups all get done 3rd.

So DRW does all in Set A first (ranks 1 3 6 12, in that order),
then all in Set B (ranks 2 7 8 9, in that order),
then all in Set C (ranks 4 5 10 11, in that order).
This means that instead of going through the ranks in the usual order 1-12,
DRW goes through them in the order 1 3 6 12 2 7 8 9 4 5 10 11.

By the time DRW finishes Set A, it has tried AA's 1-12 once each.
By the time DRW finishes Set B, it has tried AA's 2-12 twice each,
and AA's 1 & 13 once each.
By the time DRW finishes Set C, it has tried AA's 3-12 three times each,
AA's 2 & 13 twice each, and AA's 1 & 14 once each.

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