You are here

Fold tree not correct for multi-chain partners

7 posts / 0 new
Last post
Fold tree not correct for multi-chain partners
#1

I have a two partner docking problem, where the first partner is made up of chains A,B,C,and D, and the second partner is chain X. They are loaded from a single PDB.

In my script I am using:

dock_prot = DockingProtocol()
dock_prot.setup_foldtree(p,'ABCD_X')
print p.fold_tree()

Which prints:

FOLD_TREE EDGE 1 117 -1 EDGE 117 118 1 EDGE 118 227 -1 EDGE 227 235 -1 EDGE 227 479 4 EDGE 235 236 2 EDGE 236 353 -1 EDGE 353 354 3 EDGE 354 471 -1 EDGE 479 472 -1 EDGE 479 486 -1

However, once the pose is created, a command such as:

p.pdb_info().pdb2pose('X',1)

will work and returns the correct AA #. I tried creating my own fold tree, but received an error. Is my docking protocol fold tree sytax correct?

Post Situation: 
Fri, 2010-12-10 05:35
msellers

The fold tree appears to be correct. I think your confusion is caused by the fact that the fold tree reports residue numbers, not PDB numbering. (Internally, Rosetta counts residues from 1 at all times; PDB info is maintained in a separate data structure).

Is it not working...?

Fri, 2010-12-10 06:53
smlewis

The output is showing a jump at EDGE 117 118 1, but I have explicitly set it with dock_prot.setup_foldtree(p,'ABCD_X'). So the jump should be further down. Right now the docking is doing A_BCDX.

Fri, 2010-12-10 06:57
msellers

You have 4 chains, thus you have 4 jumps (A to B, B to C, C to D, and somebody to X). The issue then is telling it which jump is the MOBILE jump - in your case, 4.

Fri, 2010-12-10 06:59
smlewis

So this would be done with the dock_jump parameter?

#initial perturbation movers
randomize1 = RigidBodyRandomizeMover(p, dock_jump, partner_upstream)
randomize2 = RigidBodyRandomizeMover(p, dock_jump, partner_downstream)
dock_pert = RigidBodyPerturbMover(dock_jump, 3, 8) #3A translation, 8 degrees rotation
spin = RigidBodySpinMover( dock_jump )
slide_into_contact = DockingSlideIntoContact(dock_jump)

Fri, 2010-12-10 07:02
msellers

Dunno, but it's very likely. (I've never used PyRosetta).

Let us know if it works!

Fri, 2010-12-10 07:07
smlewis

Yup, that worked!

I needed to specify "dock_prot.setup_foldtree(p,'ABCD_X')" and then set dock_jump equal to 4.

Thanks for the help.

Fri, 2010-12-10 07:21
msellers