Hi
Follow up the shaking the coordinates, I was end up using different R-free set.

I do have a data is at very low resolution, so only possible to do rigid body refinement.

How can I  avoid over fitting or model bias due to the different R-free set using phenix.

Thanks
ram



From: Pavel Afonine <[email protected]>
To: PHENIX user mailing list <[email protected]>
Sent: Sun, March 28, 2010 10:13:08 PM
Subject: Re: [phenixbb] shaking a PDB without introducing clashes

Hi Francois,

to add to Nat's reply:

- you can also apply a systematic shift to the model by rotating+translating it:

phenix.pdbtools model.pdb rotate="10 20 30" translate="1 2 3"

- you can run SA "refinement" w/o X-ray term:

phenix.refine model.pdb fake_data.mtz simulated_annealing=true main.bulk_solvent=false main.number_of_macro_cycles=10 output.prefix=distorted_model strategy=individual_sites

My plan is to add a command line tool called phenix.fake_model_and_data that will create a model and corresponding fake data that will look like a real data and no-one could say that it is a fake data -:) To be available soon!

All the best!
Pavel.


On 3/28/10 5:17 PM, Francois Berenger wrote:
> Hello,
>
> I recently used "phenix.pdbtools modify.sites.shake=..."
> in order to have some data from 0.1 to 5A. away from
> the true structure in order to test some scripts.
>
> I think this introduced clashes in the "shaken" PDBs
> I got.
> Is there a tool/way to avoid this?
>
> Best regards,
> Francois.
> _______________________________________________
> phenixbb mailing list
> [email protected]
> http://phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________
phenixbb mailing list
[email protected]
http://phenix-online.org/mailman/listinfo/phenixbb