Recipe: 933 Secondary Structure from Contacts

created by Susume

Profile

• Name: 933 Secondary Structure from Contacts
• ID: 49562
• Shared with: Public
• Parent: None
• Children: None
• Created on: July 08, 2014 at 18:27 PM UTC
• Updated on: July 08, 2014 at 18:27 PM UTC
• Description:

  For CASP Ts802 (puzzle 933) or CASP Tc802 only! SS based on contact map. More hints about protein shape in script comments (view in script editor).

Code

--For CASP Ts802 (puzzle 933) or CASP Tc802 only
--Secondary structures based on contact map
--Helix at 59-66 may turn out to be just loop
--Overall shape seems to be a long thin barrel with two seams that don't close
--Sheet order includes a greek key (google "protein greek key")
--Some loops between sheets cross from one side of the barrel to the other

selection.SelectAll()
structure.SetSecondaryStructureSelected("L")
selection.DeselectAll()
selection.SelectRange(5,16)
selection.SelectRange(19,30)
selection.SelectRange(33,43)
selection.SelectRange(51,55)
selection.SelectRange(70,79)
selection.SelectRange(82,89)
selection.SelectRange(98,102)
selection.SelectRange(111,121)
structure.SetSecondaryStructureSelected("E")
selection.DeselectAll()
selection.SelectRange(59,66)
structure.SetSecondaryStructureSelected("H")
selection.DeselectAll()

Comments

[Bruno Kestemont]

I don't understand how you deduce structure from contact, but it seems heart and I use your structures

[LociOiling]

Contact Cement (http://fold.it/portal/recipe/49061 or http://fold.it/portal/recipe/49437) prints the contact info in spreadsheet friendly CSV format, for example:

Contact Cement v1.00
106 contacts with non-zero heat found
seg1,seg2,heat
8,27,1
9,27,1
10,25,1
12,22,1
12,23,1
12,60,1
13,21,1
…

The format is segment 1, segment 2, contact heat.

In this short example, there's segment 1:

8 - 9 - 10 - [11] - 12 - 13

and segment 2:

27 - [26] - 25 - [24] - 23 - 22 - 21.

Even though there are gaps like [11], [26], and [24], the contacts suggest two sheets in an antiparallel alignment, where the lower-numbered end of one sheet is adjacent to the higher-numbered end of the other sheet.

The contact 12,60 indicates contact with another part of the protein.

On the other hand, there's

57,61,1
58,60,1
59,61,1
59,62,1
60,62,1
61,63,1
62,64,1
63,116,1
65,68,1

Again there's an outlier at 63,116. The rest of the points have one segment in contact with another two to four segments away. A couple of points like this probably indicate a turn, but a longer list of points tends to indicate a helix.

So one technique is to paste the output from Contact Cement into a spreadsheet and manually group the contacts.