Dear all, I am trying to use the rosetta prerefinement fuction of MR_rosetta to complete a homology model with missing stretches of amino acids. Everything runs fine and residues are being added to fill the gaps using the fragment files, but rosetta is also altering parts of the model which should already be well placed. Is there a way to tell Rosetta only to prerefine certain residues? I've been following the command line suggestion: phenix.mr_rosetta \ seq_file=seq.dat \ data=coords1.mtz \ search_models=coords1.pdb \ run_prerefine=True \ number_of_prerefine_models=1000 \ fragment_files = test3.gz \ fragment_files = test9.gz \ rescore_mr.relax=False \ rosetta_models=100 \ ncs_copies=2 \ space_group=p212121 \ use_all_plausible_sg=False \ nproc=200 \ group_run_command=qsub Thanks, Heather Condurso [email protected]
Hi Heather, That would be useful, I can see that. I don't think there is an easy way to do that in mr_rosetta. There are two possibilities I can think of. 1. You can pass any Rosetta commands you want to mr_rosetta with the keyword "rosetta_command". I will ask Frank Dimaio from the Baker group about what commands might do this. 2. Alternatively you could fix it up by hand afterwards by using the phenix model tools to select some residues from the original model and some residues from the new model and then just make a composite PDB file. All the best, Tom T On Nov 8, 2013, at 11:02 AM, Heather L Condurso wrote: Dear all, I am trying to use the rosetta prerefinement fuction of MR_rosetta to complete a homology model with missing stretches of amino acids. Everything runs fine and residues are being added to fill the gaps using the fragment files, but rosetta is also altering parts of the model which should already be well placed. Is there a way to tell Rosetta only to prerefine certain residues? I've been following the command line suggestion: phenix.mr_rosetta \ seq_file=seq.dat \ data=coords1.mtz \ search_models=coords1.pdb \ run_prerefine=True \ number_of_prerefine_models=1000 \ fragment_files = test3.gz \ fragment_files = test9.gz \ rescore_mr.relax=False \ rosetta_models=100 \ ncs_copies=2 \ space_group=p212121 \ use_all_plausible_sg=False \ nproc=200 \ group_run_command=qsub Thanks, Heather Condurso [email protected]mailto:[email protected] _______________________________________________ phenixbb mailing list [email protected]mailto:[email protected] http://phenix-online.org/mailman/listinfo/phenixbb
Tom,
Thanks for the quick reply I will try #2 and see if that gives me a
reasonable model. If you hear from the Baker lab with a useful command
please let me know.
Thanks again,
~Heather
On Fri, Nov 8, 2013 at 1:26 PM, Terwilliger, Thomas C
Hi Heather,
That would be useful, I can see that. I don't think there is an easy way to do that in mr_rosetta. There are two possibilities I can think of.
1. You can pass any Rosetta commands you want to mr_rosetta with the keyword "rosetta_command". I will ask Frank Dimaio from the Baker group about what commands might do this.
2. Alternatively you could fix it up by hand afterwards by using the phenix model tools to select some residues from the original model and some residues from the new model and then just make a composite PDB file.
All the best, Tom T
On Nov 8, 2013, at 11:02 AM, Heather L Condurso wrote:
Dear all,
I am trying to use the rosetta prerefinement fuction of MR_rosetta to complete a homology model with missing stretches of amino acids. Everything runs fine and residues are being added to fill the gaps using the fragment files, but rosetta is also altering parts of the model which should already be well placed. Is there a way to tell Rosetta only to prerefine certain residues?
I've been following the command line suggestion:
phenix.mr_rosetta \ seq_file=seq.dat \ data=coords1.mtz \ search_models=coords1.pdb \ run_prerefine=True \ number_of_prerefine_models=1000 \ fragment_files = test3.gz \ fragment_files = test9.gz \ rescore_mr.relax=False \ rosetta_models=100 \ ncs_copies=2 \ space_group=p212121 \ use_all_plausible_sg=False \ nproc=200 \ group_run_command=qsub
Thanks, Heather Condurso [email protected] _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
Hi Heather,
Here is Frank DiMaio's quick reply:
Hi Tom,
There is no command line option but I can add code to do this. I can get this checked in today (and can make a patch as well).
-Frank
So once Frank does this you can download a weekly-release of Rosetta and it should then be possible, or perhaps apply the patch that he mentions if you have a very recent release of Rosetta. I'll let you know when it is available.
All the best,
Tom T
On Nov 8, 2013, at 11:37 AM, Heather L Condurso wrote:
Tom,
Thanks for the quick reply I will try #2 and see if that gives me a reasonable model. If you hear from the Baker lab with a useful command please let me know.
Thanks again,
~Heather
On Fri, Nov 8, 2013 at 1:26 PM, Terwilliger, Thomas C
Hi, my Email was registered before, and I submit my information on phenix web, then the web shows that as follow: "An e-mail message will be sent to you with a password and instructions for downloading the PHENIX software". While I can not receive the password, even in my spam folder. I try it twice, it does not work. I do not know why. Thanks! Huaidong Zhang
On Tue, Nov 12, 2013 at 4:50 AM, 张怀东
Hi, my Email was registered before, and I submit my information on phenix web, then the web shows that as follow: "An e-mail message will be sent to you with a password and instructions for downloading the PHENIX software". While I can not receive the password, even in my spam folder. I try it twice, it does not work. I do not know why.
The server log shows that the message was sent, so I have no idea where it is going, but no doubt there are other firewalls that could be blocking it. In any case I will email you the current download information off-list. Pending further debugging, it will become much easier for previously registered users to update in the very near future. -Nat
participants (4)
-
Heather L Condurso -
Nathaniel Echols -
Terwilliger, Thomas C -
张怀东