Use “loopmodel.linuxgccrelease” CCD to build missing residue positions

[Anonymous]

Dear friends,
I am trying to use the loopmodel ccd to build a loop region with missing residue positions in the original PDB file.

The whole sequence is of 228 amino acid and the missing region is 223-228.

Can I ask
1) As the loop to be built is at the end of the sequence (i.e. termini), how can I specify the loop file?

If I use “LOOP 222 228 0 0 0”, I was told (ccd_1.log attached):
“core.kinematics.FoldTree: FoldTree::reorder( 1 ) failed, new/old edge_list_ size mismatch”

If I use “LOOP 222 228 228 0 0”, I was told the same thing (ccd_2.log attached).

It is exciplitly indicated that “quick_ccd can also remodel termini”
https://www.rosettacommons.org/docs/latest/loopmodel-ccd.html
So how can I really do this?

2) I am building homology model. I would like to minimize the relax in the rest of the structure (i.e. before residue 223).

I am going to use the options below following “bcorreia” said in
https://www.rosettacommons.org/node/2248

Is this suitable?

-nstruct 1000
-s path/to/pdb
loops:loop_file path/to/loop_file
-loops:relax fastrelax
-relax::fastrelax_repeats 8
# NOT use -loops:extended
-loops:frag_sizes 9 3 1
-loops:frag_files path/to/frag_files
-loops:remodel no
-loops:refine refine_ccd
-out:file:silent_struct_type binary
-out:file:silent path/to/silent_file
-out:file:fullatom

3) Does “-relax::fastrelax_repeats” exist in the “loopmodel.linuxgccrelease”?
I was told

ERROR: ERROR: Option matching -relax:fastrelax_repeats not found in command line top-level context

when using it.

Thank you very much.

Yours sincerely
Cheng

[Anonymous]

There is at least one demo on loop length changing. The one I know of is at Rosetta/demos/protocol_capture/2010/AnchoredDesign/loop_length_changing. This demo assumes you have an internal loop, but the technique will probably work for termini.

The key is that Rosetta is not doing the extension. Give Rosetta a PDB with the correct number of residues. Generate that by adding residues in a text editor. Garbage coordinates are ok so long as the backbone heavyatoms (C, CA, N, O) are present. Rosetta has occasional bugs with too many (0,0,0) atoms, so you may want to slightly randomize the coordinates, or just copy coordinates from elsewhere in the protein. Make sure you are using the “extended loop” option in the loop file and it will probably work.

The name of the fastrelax option appears to have been changed to relax:default_repeats.

[Anonymous]

Hi smlewis,
Thank you for your help. I will try your methods later. In the meantime, I have been trying to use “floppy tail” as I think the scenario is exactly what I want.

https://www.rosettacommons.org/docs/latest/floppy-tail.html

I will ask the question in another post.

Yours sincerely
Cheng

[Author]

Posts