# ddG = dGmutant ‐ dGwildtype

4 posts / 0 new
ddG = dGmutant ‐ dGwildtype
#1

Hello,
I would like to ask you about computations of a change of stability, 'ddG = dGmutant ‐ dGwildtype'. I use Rosetta for the very first time. I employ Rosetta 3.9.
I have adjusted our crystal structure 'structure.pdb' by MolProbity (addition of hydrogens) and by the Rosetta program 'clean_pdb.py'. The file structure.pdb is not minimized/optimized/relaxed.

I have prepared input.mutfile for  30 single mutations in this way:
total 30
1
R 60 Q
1
E 62 K
...
but I am not sure whether it is correct.

'High-Resolution Protocol Flags' is convenient for my purpose, therefore, I have copied the options from https://www.rosettacommons.org/docs/latest/application_documentation/analysis/ddg-monomer to a flags file.
Unfortunately, I do not understand how to set some options.

According to comments on the options, I think that minimization can run by these default settings without other keywords:
-ddg:minimization_scorefunction
-ddg::minimization_patch
Is it correct?

A similar problem lies in a preparation of a constraint file.  According to the section Algorithm (the link above), it is possible to choose an automatically generated file, which I prefer, but I do not know how to set it. If the keyword 'ddg::min_cst true' automatically makes the constraint file does it mean that the keyword is only '-constraints::cst_file' so without a name specification?

And my last question is how to run it in the command line with and without flags file?  I have found this link https://www.rosettacommons.org/manuals/rosetta3.1_user_guide/app_ddg_prediction.html where all options are explicitly put in the command line but I do not understand how to work with flags.

Would you be so kind as to help me, please?

Ela

My Flags:
in:file:s /path/structure.pdb
-ddg::mut_file /path/input.mutfile
-ddg:weight_file soft_rep_design
-ddg:minimization_scorefunction
-ddg::minimization_patch
-database /software/rosetta/3.9/main/database
-fa_max_dis 9.0
-ddg::iterations 50
-ddg::dump_pdbs true
-ignore_unrecognized_res
-ddg::local_opt_only false
-ddg::min_cst true
-constraints::cst_file
-ddg::suppress_checkpointing true
-in::file::fullatom
-ddg::mean false
-ddg::min true
-ddg::sc_min_only false
-ddg::ramp_repulsive true
-mute all
-unmute core.optimization.LineMinimizer
-ddg::output_silent true

Post Situation:
Thu, 2018-07-26 09:17
ela

You probably want to use one of the benchmarked sets of options, rather than rolling your own. The two options sets on the documentation page ("row 16" and "row 3") are your best bets.

If you include -ddg:minimization_scorefunction or -ddg::minimization_patch, you need to provide a value to them. When it mentions they're optional, that means you can just delete those lines from the options file entirely.  (By the way, the default scorefunction in Rosetta has changed since the protocol has originally been published. If you omit the scorefunction now with 3.9, you'll get REF2015-based results, rather than score12-based results that were in the Kellogg paper. We think the new energy function should be better, but if you want to return to older results, you need to add the option -restore_pre_talaris_2013_behavior to the commandline.)

To use the automatically generated constraints, simply omit the -constraints::cst_file option. Rosetta will recognize that that option is not being passed, and automatically generate the constraints internally. (Omit entirely - if you use that option you need to pass a filename to it.)

To use an option file, simply pass it on the commandline with a '@' before it.  So something like /path/to/Rosetta/main/source/bin/ddg_monomer.linuxgccrelease @options_file.txt  -- but substitute the name of your options file to options_file.txt. Note that you can mix commandline and options files. I'd actually recommend puling the -in:file:s -ddg::mut_file out of the options file and put it on the commandline, so you have a general options file which can be re-used on multiple different proteins. e.g. /path/to/Rosetta/main/source/bin/ddg_monomer.linuxgccrelease @options_file.txt -in:file:s /path/structure1.pdb -ddg::mut_file /path/struct1.mutfile`

Another thing to be aware of is that there's a new ddG protocol out there, cartesian_ddg which is showing better results in certain situations than the original Kellogg et all protocol. See https://www.rosettacommons.org/docs/latest/cartesian-ddG and references therein.

Fri, 2018-08-03 09:23
rmoretti

Hello,

Thank you for your response.  I kept your recommendations. Unfortunately,  I did not get the correct outputs.  The input files and the incorrect outputs are attached.

I used the procedure as follows.

1.  I had had a problem with my PDB file, therefore I adjusted the format by a recommendation that I found in another forum discussion.
/software/rosetta/3.9/main/source/bin/score_jd2.mpi.linuxiccrelease -in:file:s dep2.pdb -out:pdb -out:file:renumber_pdb
It worked.

2.   I used the minimize procedure as it is described in the tutorial here https://www.rosettacommons.org/docs/latest/application_documentation/analysis/ddg-monomer
The constraints file was generated as well.

/software/rosetta/3.9/main/source/bin/minimize_with_cst.mpi.linuxiccrelease -in:file:l list -in:file:fullatom -ignore_unrecognized_res -fa_max_dis 9.0 -database /software/rosetta/3.9/main/database/ -ddg::harmonic_ca_tether 0.5 -score:weights score12 -ddg::constraint_weight 1.0 -ddg::out_pdb_prefix min_cst_0.5 -ddg::sc_min_only false -score:patch /software/rosetta/3.9/main/database/scoring/weights/score12.wts_patch > mincst.log

/software/rosetta/3.9/main/source/src/apps/public/ddg/convert_to_cst_file.sh mincst.log > input.cst

3.   Finally, I used the 'High Resolution Protocol'  (see the link above):

/software/rosetta/3.9/main/source/bin/ddg_monomer.mpi.linuxiccrelease @options_file.txt

According to information in the section 'Expected Outputs' (see the put link above), I should obtain ddg_predictions.out  and wt_<WT_AA><RESIDUE_NUM><MUTANT_AA>.out and mut_<WT_AA><RESIDUE_NUM><MUTANT_AA>.out but I did not.
Into the options file, I put -ddg::out ddg_predictions.out  and -ddg::output_silent true
I got the only file wt_.out which is incorrect. I tried it several times but without success.  The incorrect outputs are only for 2 single mutations and 10 interactions (the testing version).

Would you be so kind as to help me again, please?