Blueprint Tool question

Case number:699969-2003478
Opened by:gitwut
Opened on:Thursday, February 23, 2017 - 14:51
Last modified:Saturday, February 25, 2017 - 03:06

I don't understand Rama Maps or the ABEGO lingo (BAAB), so I can't frame this question in a very scientific manner. Hopefully, it won't be too hard to figure out what I'm asking.

In the attached image, I've included a screen shot of my 1342 Blueprint tool. My 47-51 sheet-helix loop passes the Ideal loop filter and I have the full 810 point bonus, but it is not in the list of suggested patterns. I understand that not all possible loops are included in the library, but I have encountered this same problem (possibly a different loop pattern) on one or two of the other design puzzles.

I refer to the patterns by color, so the color for this loop is "bbrbr".

Is this a problem similar to what bkoep mentioned in the blog about the Blueprint tool (BAAB)? I removed the constraints early in the puzzle (I believe it started out as the pattern "bgbbr"). If it is an undesirable pattern, wouldn't it be scoring much lower on the Remix tool? Not sure if it matters, but Auto Structures matches the Blueprint tool SS displayed.

foldit_1487853377.png141.66 KB
(Thu, 02/23/2017 - 14:51  |  2 comments)

Joined: 04/24/2014
Groups: None

I've pinged bkoep to see if he has any insight for you. :)

bkoep's picture
User offline. Last seen 1 hour 20 min ago. Offline
Joined: 11/15/2012
Groups: Foldit Staff

Good question, gitwut! In Puzzle 1342 (and most recent design puzzles), the Ideal Loop Filter is programmed to be a little tolerant about where sheets begin and end. As I'm sure you know, sheet detection by Auto Structures can be a little finicky; a residue can have a sheet-like backbone (ABEGO blue), but is only recognized as a sheet if it makes hydrogen bonds with another sheet. So, if the Ideal Loop Filter checks your loop and doesn't see a sheet where it expects a sheet, the filter will check to see if the sheet starts at the next residue.

In this case, the filter recognizes the Sheet-Helix BABA (or "brbr") Ideal Loop, and expects to see a sheet at residue 48. Instead, it sees that residue 48 has a sheet-like backbone, but isn't making the necessary hydrogen bonds to be considered a sheet. Due to the tolerance settings, the filter gives your loop the benefit of the doubt and checks to see if residue 47 is part of sheet. Residue 47 is indeed part of a sheet, so the filter passes.

This "tolerance" is actually apparent in the original study that identified the Ideal Loops. And we've seen a lot of good-looking designs from these puzzles!


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, Boehringer Ingelheim, RosettaCommons