How do I explicitly add residues for design in the interface optimization script?

[Anonymous]

I am playing around with the interface optimization script in the demo “design_raf_rac_interface” (rosetta3.5). For the system I am working with, I would like to add a few residues to the list of residues to be redesigned, an not redesign all residues of the interface, but only those around a mutated residue in the binding partner.

For the script used in the demo,these residues are repacked and redesigned, based on the ProteinInterfacDesign interface_distance_cutoff=8:
PackingRes, 1 B, 2 B, 3 B, 4 B, 5 B, 6 B, 7 B, 8 B, 9 B, 10 B
DesignRes, 36 A, 39 A, 40 A, 75 A, 78 A, 79 A, 82 A, 117 A, 120 A, 121 A, 124 A, 159 A, 162 A, 163 A, 166 A, 201 A, 204 A, 205 A, 208 A, 243 A, 246 A, 247 A, 250 A

(only 3 residues out of 7 potential contact residues in the pocket were actually redesigned)

I would like to redesign only residues 155 A , 159 A , 162 A , 194 A , 197 A , 198 A , 201 A, and repack all interface residues and some additional residues around the additionally redesigned residues (those are the residues which we randomized in the actual wet-lab experiments)

Taking this script as an example (https://www.rosettacommons.org/sites/default/files/design_script.xml_.txt), I added
a res.file to the script:

under :

defines the residue file to be used and reads it into variable inter_mut.

under : passes variable inter_mut to the PackRotamer mover

The residue file is read and applied, however, now these residues are repacked and redesigned:

PackingRes, 36 A, 40 A, 75 A, 78 A, 79 A, 82 A, 117 A, 120 A, 121 A, 124 A, 163 A, 166 A, 204 A, 205 A, 208 A, 243 A, 246 A, 247 A, 250 A, 285 A, 289 A,
1 B, 2 B, 3 B, 4 B, 5 B, 6 B, 7 B, 8 B, 9 B, 10 B
DesignRes, 159 A, 162 A, 201 A

That is, only those residues that are both in the interface and in the residue file get redesigned, those that are outside the interface are ignored.

How do I alter the script to gain control over which residues are only repacked, and which get redesigned?

[Anonymous]

This is one of the “gotchas” for PackerTask (TaskOperation) specification: various operations combine to give the most restrictive behavior. That is, instead of specifying design on particular residues, you’re more properly specifying that all the other residues aren’t being designed.

Only residues which are allowed to design by all task operations (159 A , 162 A , 201 A) actually get redesigned, and the the others get the more restrictive “repack only” or “no repack” option that the other taskoperation specifies.

My suggestion is, given you know exactly which residues you want to repack and which you want to redesign, just to completely specify the repacking and design behavior with a resfile, and just remove the ProteinInterfaceDesign/design task operation. That task operation is valuable if you want to autodetect the interface residues, but if you have a more complicated interface definition, it’s better to specify it manually. (Note that for 3.5 you can specify behavior for chain and residue ranges, which may make your resfiles easier to write: https://www.rosettacommons.org/manuals/archive/rosetta3.5_user_guide/d1/d97/resfiles.html )

[Anonymous]

Thank you very much, that seems to work.

[Author]

Posts