You are here

RosettaScripts - FastRelax loops

3 posts / 0 new
Last post
RosettaScripts - FastRelax loops

Hi all,

I am using RosettaScripts for some loop modelling and would like to apply the FastRelax protocol to a loop region in my structure without the moves propagating down the chain.

I know that it is possible to set up a MoveMap so that only some degrees of freedom are turned on e.g.

                <FastRelax name=frel scorefxn=SFXN_relax repeats=8 >
                                <Chain number=1 chi=0 bb=0 /> Turns off all degrees of freedom for chain A
                                <Span begin=69 end=76 chi=1 bb=1/> This line turns the side-chain (chi) and backbone (bb) freedoms back on

however this propagates moves down the chain so that it is not a local move. I tried adding a chainbreak to my protocol but this does not seem to behave as expected (I've probably misunderstood something). 

Is there a simple way of applying FastRelax (in both dihedral and cartesian space) to a local loop region? I've noticed there is a LoopRelaxMover but this is not documented so I'm not clear on how to use it.

Any help would be greatly appreciated!






Post Situation: 
Wed, 2015-11-18 09:32

The issue here is that you need to set up the FoldTree such that you have a chainbreak within the loop, and a jump across the loop. This would mean that any changes to the loop backbone dihedrals propagate toward the cutpoint, and the non-loop regions are not connected to the root of the FoldTree by anything that moves during relax.

I believe the FoldTreeFromLoops mover ( is what you want. If you apply it before the relax mover, it should set up the appropriate loops FoldTree, which the Relax mover should obey..

The only wrinkle is that the scorefunction you use for the FastRelax would need to have the "chainbreak" scoreterm turned on (typical value is 1.0). This is what keeps the cutpoint from opening up. Without it, the cutpoint will be an actual break in the backbone.  The default talaris2013 scorefunction does not have this, so you may need to use the Reweight facility to make a custom scorefunction with the term on. (


Tue, 2015-11-24 15:38


Thanks a lot for your help! 

I'll give this a try.


Fri, 2015-12-04 03:11