Re: [phenixbb] Request for C shell script file for phenix.refine

Hi Partha,

Thanks a lot for your suggestions. Its close but not completely happy yet. So, I generated the file as you suggested, several of the commands are understandable, and I can run it after modification as:

$ phenix.refine

Loose the ' < '

but would be easier if there is a scrip for phenix.refine as available for SOLVE/RESOLVE. Could you point me to the not so easy way of doing that?

I must confess that I think that the phenix.refine interface is as easy as it gets and is a much better option than tradiaionl input files. If you have the feeling that the parameter file is too overwhelming in terms of number of parameters, just loose the scopes and parameters you are not interested in: default values will be used in those cases. If you really want to go for keyword based input as is done in SOLVE/RESOLVE , I recommend you read the newsletter http://cci.lbl.gov/publications/download/iucrcompcomm_jan2005.pdf on the PHIL parameter file.

One particular problem seems to be modified residues, in my case, a phosphothr (TPO), had to delete the phosphate and rename TPO as THR.. is there any wayout?

Just read in the a cif file like this phenix.refine m.mtz m.pdb tpo.cif or like this phenix.refine m.def tpo.cif or add it in the parameter file in the appropriate scope. HTH P

Best, Partha

On 9/28/07, Peter Zwart wrote:

...

The easiest way is to have a parameter file that you edit. The easiest way to get a parameter file is by typing

phenix.refine <model> <data> --dry_run

and use the subsequently generated ???_refine_001.eff file.

rename that eff file to something like

parameters.def

which you can use for subsequent customisations and editing of parameters.

Not ethat every time you finish a refinement job, a new def file is generated of rthe next round, updating the model info for further rounds of refinement.

P

2007/9/28, Partha Chakrabarti < [email protected]>:

...

Hi,

I am wondering if one can have master scripts (C shell) for the macro cycle of phenix.refine that can be used from the command line. It is easier to comment out an option than having to type the whole thing.. has anybody tried it? Similar scripts exist for solve/resolve and phaser..

I should not say it, I am myself not adverse to using the GUI, it is just easier when the poor postdoc sits with the boss.. typically I face the question: where is the script file & where is the log file :))

Cheers, Partha

-- MRC National Institute for Medical Research Division of Molecular Structure The Ridgeway, NW7 1AA, UK Email: [email protected] Phone: + 44 208 816 2515 _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

--

MRC National Institute for Medical Research Division of Molecular Structure The Ridgeway, NW7 1AA, UK Email: [email protected] Phone: + 44 208 816 2515 _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb