4 replies [Last post]
Susume's picture
User offline. Last seen 1 day 56 min ago. Offline
Joined: 10/02/2011

I found today that placing a cut in an ideal loop causes it to become non-ideal, and adds a -100 penalty to the ideal loop filter. Placing a cut in a non-ideal loop does not increase the filter penalty. Placing a cut in a sheet or helix does not cause a score penalty. Placing a cut between a sheet or helix and an ideal loop does cause the score penalty. All of this could be used to check in a script if a loop is ideal or not.

Manually set all the SS before running script. Check score with filters on and filters off - difference is filter score. Find a loop. Place a cut in it. Check score with filters on and off. Did filter score go down by 100? If yes, the loop was already ideal - undo the cut. If no, the loop either a) was not ideal and needs to be worked on, or b) is currently considered to be part of a sheet or helix. Not sure how to tell these two conditions apart.

Joined: 09/24/2012
Groups: Go Science
Good idea

This is a "idealitycheck" function that could be added in recipes (like remix recipes).

I found that an ideal loop might change status if the sheet bonds change. For example, a loop is recognized as loop and it is ideal. Then a bond is broken and the first sheet segment is seen as a loop, the loop turns unideal. This idealitycheck might be used regularilly to select appropriate actions (either return to previous solution with bonds, either re-work on this loop).

Joined: 09/24/2012
Groups: Go Science
To do this

I suggested a new argument "A" ("auto") for Lua SetSecondaryStructure functions here:
http://fold.it/portal/node/2002998

The problem beeing that with user's SS, this idea doesn't work well. It works only when the loop is considered a loop by Foldit.

Susume's picture
User offline. Last seen 1 day 56 min ago. Offline
Joined: 10/02/2011
A Lua call to set SS to what

A Lua call to set SS to what foldit thinks it is is a great idea. I have my doubts about the current auto structures tool, though - I don't think it uses the same criteria as the ideal loop filter. If the "auto" parameter on the Lua call could be made to use the criteria that the ideal loop filter uses, this would really help with scripting.

Joined: 09/24/2012
Groups: Go Science
VoilĂ 

Here a recipe that uses this feature:
http://fold.it/portal/recipe/102113

Note that it can only discriminate Foldit ideal loops (or sheets or helices) from Foldit non ideal ones. It can not discriminate what you consider a loop or not yourself.

To see what Foldit considers a loop, sheet or helice, do "auto-structure".

A good way to tell the recipe which loop is unideal is to change the entire segment the filter considers unideal (with red spots) to loop. You can keep your other loops as you want.

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