Back to Recipes Homepage
recipe picture
Recipe: TvdL DRemixW 3.1.0 Orange Guard
Created by Susume 102 241
4.8
Your rating: None Average: 4.8 (10 votes)

Profile

Name: TvdL DRemixW 3.1.0 Orange Guard
ID: 103398
Created on: Sun, 04/12/2020 - 10:15
Updated on: Sun, 04/12/2020 - 17:15
Description:

Freeze sidechains before starting that you want to be kept orange. It will only mutate those to orange sidechains; everything else mutates freely.



Best For


Comments

Susume's picture
User offline. Last seen 15 hours 52 min ago. Offline
Joined: 10/02/2011
Orange Guard is for Science

Orange Guard lets you protect orange sidechains that foldit wants to turn blue, by freezing those sidechains before you start the script. It will try to mutate those sidechains, but only to other orange ones.

Orange Guard will *not* score better than regular DRemixW, since it limits the mutations that foldit is allowed to try. The more sidechains you protect by freezing them, the more the score is limited. So why run it? It prevents foldit from producing a protein with a long stretch of only blue sidechains (say, an entire edge sheet, or more than one turn of a helix). The scientists generally don't pick a protein for testing in Rosetta if it has an entire blue sheet or a long stretch of blue helix, because they know the protein will not fold up as designed.

Use Orange Guard if getting your protein tested is more important to you than getting the top score. Better still, run Orange Guard in one track and the regular script in another track (you can use the same script, just don't freeze any sidechains first). Then you can have the best of both worlds.

Joined: 09/29/2016
Groups: Gargleblasters
Bug Report?

Alright so after looking at the code through my untrained eyes, I'm now uncertain if this is even a bug due to your Comment of "Used only during Initialization - after that use AllowedList instead."
This made me second guess everything, considering I lack a pre-Recipe save to compare this 16hrs-later result to. As originally for all I know, that Valine may have been a -phyllic. In which case, things have obeyed the code and there's no bug. That being said, I'm still leaving my original "report" below jsut in case.

On that note, I DIO have a request, if possible and if it wouldn't cause an epic slowdown in the recipe:
If an AA is changed to -phobic and there is a Gain, then re-run the Initialization so that the newly-set -phobic can also be protected!
Ideally a tickbox option so that one can use either-or method. :)

Either way, thanks again for these OG recipe variants!






[Original Message]
Just happened to stop its run after the last Fuze was close to my Best, as I was curious to see what the exact differences were...

The biggest was the 2 mutations, and one of them was a Threonine which originally had been a Valine... So it didn't properly Guard that Orange. :(

I can't say for certain if any other -phobics were mutated into -phyllics, but hopefully not.

My guess was an error in your tables; however, as I look at the code it SEEMS like that's all correct. (from my not-a-coder perspective)

Joined: 09/29/2016
Groups: Gargleblasters
You know what...... Ignore me! lmao

I completely missed the part where you're supposed to FREEZE what you want to not be mutated! Was 2am when I set that up last night and I didn't realize it till reading the description again just now.

My suggestion DOES still stand, but I guess it'd also need to include the ability to auto detect all -phobics and not mutate them ever OR somehow remember the originally selected. Which sounds really involved on the first part, and probably not doable with Foldit's scripting engine for the OR part...

Sorry for my braindead moment!

Joined: 09/29/2016
Groups: Gargleblasters
Real issue this time! Scouts honor... lol

The recipe also unfreezes the backbone when unfreezing the AA, instead of only the AA.

I had wanted to keep the structure of certain parts, to prevent remix and wiggling, but allow its AA to still be mutated to something still-Orange. Alas, now half of the backbones are also unfrozen :P

Joined: 06/06/2013
Groups: Gargleblasters
Formula question

are you letting the recipe run in loop mode? I often let this default setting run. I haven't tried this yet... but double check. Otherwise, need to freeze SS and sidechain both to keep everything

Joined: 09/29/2016
Groups: Gargleblasters
Better late, than never!

Sure it's been a few months but...... Shh :P

I'm just going off what Susume's instructions are in her first post:
you protect orange sidechains that foldit wants to turn blue, by freezing those sidechains before you start the script.

As for "loop mode", no clue what you're referring to by that.
Pretty much all I do on the DRW/DRemixW recipes like this one, is to change the Wiggle Factor plus the "Min" and "Max" lengths, opting to leave anything else (that I don't generally understand the functionality of) to their default values. Since I typically don't let a recipe run on end for days, they've never been allowed to "stop" on their own. Most DRW-flavor recipes I use manage at least 3 or 4 complete cycles before being stopped (ie roughly 8 to 12 hours; longer if it's gained every length in a cycle)

Ooooh you mean "Convert to Loops". Yes, do I generally use that. :P
But as I said, I had wanted to prevent it from running Remix/Rebuild on certain segments, so I froze their SS to keep them from having Remix or Rebuild run on them. However, the recipe unfreezes everything after it initializes, so it doesn't matter what you've frozen. It keeps track of the frozen Sidechains, in order to mutate them to only Phobics (as quoted above). Don't know why it unfreezes my Backbone segments though. Hence why I reported it as a bug :)

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