Dear G, after carefully re-reading your email, I think I have only given half of the answer. By repeating the 'coordinates' scope, you are defining one ensemble using multiple PDB files. If you want to define multiple ensembles, you need to repeat the whole ensemble scope, i.e. ensemble { model_id = a1 coordinates { pdb = model1.pdb identity = 0.9 } } ensemble { model_id = a2 <= note the change in model_id coordinates { pdb = model2.pdb identity = 0.9 } } If you want Phaser to actually search for all these ensembles, you need to add a search keyword as well: 1. Conduct a separate search for each ensemble listed. The search order is decided automatically. The solution contains one molecule of a1 + one molecule of a2. search { ensembles = a1 copies = 1 } search { ensembles = a2 copies = 1 } 2. Phaser will search for each alternative ensembles in turn and pick the best one (the solution will contain one molecule of a1 or one molecule of a2, but not both): search { ensembles = a1,a2 copies = 1 } HTH, Gabor On Aug 8 2011, Dr G. Bunkoczi wrote:

Dear G,

I think 'coordinates' is a multiple scope, so you have to define your ensemble as follows:

ensemble { model_id = a1 coordinates { pdb = model1.pdb identity = 0.9 } coordinates { pdb = model2.pdb identity = 0.9 } coordinates { pdb = model3.pdb identity = 0.9 } coordinates { pdb = model4.pdb identity = 0.9 } }

BW, Gabor

On Aug 8 2011, G Y wrote:

...

Dear all,

I am writing a bash script for running phenix.phaser automatically on an ensemble of multiple pdbs. These pdbs have been superimposed by other tools before running phaser.

The following is what I wrote for the ensemble part in params.eff file:

ensemble { model_id = a1 coordinates { pdb = model1.pdb pdb = model2.pdb pdb = model3.pdb pdb = model4.pdb rmsd = None identity = 0.9 }

It runs well. When it finished, I want to confirm that it really run MR on all four models or on a ensemble of these models. However in .log file, I found at the refinement part it shown me:

---------- REFINEMENT ----------

There is 1 solution to refine Refining solutions 0% 100% |==| DONE

Configuring ensembles --------------------- PDB file # 1: model1.pdb This pdb file contains 1 models The (input) RmsD of this model with respect to the real structure is 0.8

**************************** Does this mean that phenix.phaser only run MR with the first model? Does it due to the way I wrote in my params.eff file? What would be the correct way?

Many thanks!

best, G

-- ################################################## Dr Gabor Bunkoczi Cambridge Institute for Medical Research Wellcome Trust/MRC Building Addenbrooke's Hospital Hills Road Cambridge CB2 0XY ##################################################

Dear Gabor, Thanks a lot for your detailed explanation and kind help. I tried your comment. The two ways of defining search procedure gave me the same solution. For search { ensembles = a1 copies = 1 } search { ensembles = a2 copies = 1 } I got also only one result contains one best molecule pdb. Would you have some idea for this? How could I really get several output for each of the input model? Thanks and regards, G On Mon, Aug 8, 2011 at 3:13 PM, Dr G. Bunkoczi wrote:

Dear G,

after carefully re-reading your email, I think I have only given half of the answer. By repeating the 'coordinates' scope, you are defining one ensemble using multiple PDB files. If you want to define multiple ensembles, you need to repeat the whole ensemble scope, i.e.

ensemble { model_id = a1 coordinates { pdb = model1.pdb identity = 0.9 } } ensemble { model_id = a2 <= note the change in model_id

coordinates { pdb = model2.pdb identity = 0.9 } }

If you want Phaser to actually search for all these ensembles, you need to add a search keyword as well:

1. Conduct a separate search for each ensemble listed. The search order is decided automatically. The solution contains one molecule of a1 + one molecule of a2.

search { ensembles = a1 copies = 1 } search { ensembles = a2 copies = 1 }

2. Phaser will search for each alternative ensembles in turn and pick the best one (the solution will contain one molecule of a1 or one molecule of a2, but not both):

search { ensembles = a1,a2 copies = 1 }

HTH, Gabor

On Aug 8 2011, Dr G. Bunkoczi wrote:

Dear G,

...

I think 'coordinates' is a multiple scope, so you have to define your ensemble as follows:

ensemble { model_id = a1 coordinates { pdb = model1.pdb identity = 0.9 } coordinates { pdb = model2.pdb identity = 0.9 } coordinates { pdb = model3.pdb identity = 0.9 } coordinates { pdb = model4.pdb identity = 0.9 } }

BW, Gabor

On Aug 8 2011, G Y wrote:

Dear all,

...

I am writing a bash script for running phenix.phaser automatically on an ensemble of multiple pdbs. These pdbs have been superimposed by other tools before running phaser.

The following is what I wrote for the ensemble part in params.eff file:

ensemble { model_id = a1 coordinates { pdb = model1.pdb pdb = model2.pdb pdb = model3.pdb pdb = model4.pdb rmsd = None identity = 0.9 }

It runs well. When it finished, I want to confirm that it really run MR on all four models or on a ensemble of these models. However in .log file, I found at the refinement part it shown me:

---------- REFINEMENT ----------

There is 1 solution to refine Refining solutions 0% 100% |==| DONE

Configuring ensembles --------------------- PDB file # 1: model1.pdb This pdb file contains 1 models The (input) RmsD of this model with respect to the real structure is 0.8

**************************** Does this mean that phenix.phaser only run MR with the first model? Does it due to the way I wrote in my params.eff file? What would be the correct way?

Many thanks!

best, G

-- ##############################**####################

Dr Gabor Bunkoczi

Cambridge Institute for Medical Research Wellcome Trust/MRC Building Addenbrooke's Hospital Hills Road Cambridge CB2 0XY ##############################**####################

______________________________**_________________ phenixbb mailing list [email protected] http://phenix-online.org/**mailman/listinfo/phenixbbhttp://phenix-online.org/mailman/listinfo/phenixbb