Loop modeling with multiple chains

[Anonymous]

Hi,

Is it possible in rosetta to model loops for a multi chain protein.

This is the command I run for a single c hain protein:

rosetta.gcc ee 3bf2 _ -relax -looprlx -s 3bf3.pdb -loop_file 3bf2_.loops -loop_model -nstruct 5 > ee3bf3.log

This works. Now, if I have multiple c hains, can I do this:

rosetta.gcc ee 3bf2 _ -relax -looprlx -s 3bf3.pdb -loop_file 3bf2_.loops -loop_model -read_all_chains -nstruct 5 > ee3bf3.log

I tried it, and in the output file while making the first structure, it outputted, indefinetly many times:

1069900000 tries in choose_fragment_pose

The number in the above “error” kept increasing. It never stops doing this till I hit ctrl-C.

Is is possible to do mulitple chains? Am I using the proper syntax? Do I need to remove chain IDs, TER commands and/or start residues/atoms from 1?
My fasta file used to make fragments is the entire sequence, with nothing to seperate the chains.

Thanks,
Sonya

[Anonymous]

Loop modeling needs to read in fragment file.Did you change the fragment file when you switched from one chain to multi-chains?Because, obviously the sequence will be different in this case.

> Hi,
>
> Is it possible in rosetta to model loops for a multi chain protein.
>
> This is the command I run for a single c hain protein:
>
> rosetta.gcc ee 3bf2 _ -relax -looprlx -s 3bf3.pdb -loop_file 3bf2_.loops -loop_model -nstruct 5 > ee3bf3.log
>
> This works. Now, if I have multiple c hains, can I do this:
>
> rosetta.gcc ee 3bf2 _ -relax -looprlx -s 3bf3.pdb -loop_file 3bf2_.loops -loop_model -read_all_chains -nstruct 5 > ee3bf3.log
>
> I tried it, and in the output file while making the first structure, it outputted, indefinetly many times:
>
> 1069900000 tries in choose_fragment_pose
>
> The number in the above “error” kept increasing. It never stops doing this till I hit ctrl-C.
>
> Is is possible to do mulitple chains? Am I using the proper syntax? Do I need to remove chain IDs, TER commands and/or start residues/atoms from 1?
> My fasta file used to make fragments is the entire sequence, with nothing to seperate the chains.
>
> Thanks,
> Sonya

[Anonymous]

Yeah i redid the fragment files. The command line I gave for single chain was just an example of what syntax I used. they are not actually what I entered for single chain. I did single chain modeling weeks ago, so I don’t have my commands easily accesible, however I can look it up if it’ll help. The commands I gave are what I used for multi chain.

Thanks for your help.

Sonya

[Anonymous]

I got it to work with multiple chains, with only one region in the loops file instead of two as I was trying earlier. My proteins a homodimer with four chains (chain A&B make one monomer, chain C&D make the other). I was trying earlier to sample two identical regions, one in B and the in D. This was giving the error I mentioned. I broke it down to just one monomer, with two instead of 4 chains and sampling only one region. This worked fine with the syntax I mentioned earlier.
This should work enough for me, although the error does seem odd.

Thanks for looking at my problem.
Sonya

> Hi,
>
> Is it possible in rosetta to model loops for a multi chain protein.
>
> This is the command I run for a single c hain protein:
>
> rosetta.gcc ee 3bf2 _ -relax -looprlx -s 3bf3.pdb -loop_file 3bf2_.loops -loop_model -nstruct 5 > ee3bf3.log
>
> This works. Now, if I have multiple c hains, can I do this:
>
> rosetta.gcc ee 3bf2 _ -relax -looprlx -s 3bf3.pdb -loop_file 3bf2_.loops -loop_model -read_all_chains -nstruct 5 > ee3bf3.log
>
> I tried it, and in the output file while making the first structure, it outputted, indefinetly many times:
>
> 1069900000 tries in choose_fragment_pose
>
> The number in the above “error” kept increasing. It never stops doing this till I hit ctrl-C.
>
> Is is possible to do mulitple chains? Am I using the proper syntax? Do I need to remove chain IDs, TER commands and/or start residues/atoms from 1?
> My fasta file used to make fragments is the entire sequence, with nothing to seperate the chains.
>
> Thanks,
> Sonya

[Author]

Posts