[phenixbb] mr_rosetta with fixed solutions for other chains

[Thomas C. Terwilliger]

Hi Joel,

Not exactly...mr_rosetta can only run the MR portion on one component.

However in your case you can break it down.  First you can run rosetta on
your search model for the 3rd component prior to MR. Then you can run MR
with phaser in the usual way with the rosetta model for the 3rd component.
Then you can feed the entire MR solution into mr_rosetta and have it
rebuilt.

For the first step above you can use:


phenix.mr_rosetta \
  seq_file=seq.dat \
  search_models=coords1.pdb \
  run_prerefine=True \
  number_of_prerefine_models=1

Your pre-refined model(s) will be listed in

MR_ROSETTA_1/GROUP_OF_PLACE_MODEL_1/RUN_FILE_1.log

and you can pick the best of these (most negative score, listed first).

I hope that helps!

All the best,
Tom T

>> Hi-
>>
>>
>>
>> We have data for a crystal where two of three components have high
>> homology search models and can be found with straightforward phaser runs
>> but the third component has much weaker homology to anything in the pdb.
>> Is there a way to provide the solution for the first two components and
>> let mr_rosetta work on the third?
>>
>>
>>
>> Thanks,
>>
>>
>>
>> Joel
>>
>>
>>
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb at phenix-online.org
>> http://phenix-online.org/mailman/listinfo/phenixbb
>>