hi everyone: Thanks for all the response! what I said last time may be chinglish and lead to some misunderstanding, the word I said fix a helix means that the model quality of the helix is not very good, I want to modify it manually.It is so hard to do this in low resolution.I modified the helix according to the electron density map, after refinement, large red areas appear in the difference map. That made me crazy! Are there any suggestions or any reviews about this point? I am new beginner to structural biology, any suggestions are welcomed ! Best regards Xinghua Qin Xinghua Qin College of Biological Sciences No.2, Yuan Ming Yuan West Road Haidian District?Beijing?China?100094 Tel: +86-10-62732672

Message: 2 Date: Fri, 16 Jul 2010 22:10:32 -0700 From: Pavel Afonine To: PHENIX user mailing list Subject: Re: [phenixbb] sum(occ)<1? Message-ID: <[email protected]> Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi Ed,

...

Reset occupancies that are outside the [0,1] range at every step of minimization.

this simply will not work: this will invalidate the internal memory of LBFGS minimizer that we use in phenix.refine and as result it will drive it crazy.

Pavel.

------------------------------

Message: 3 Date: Fri, 16 Jul 2010 22:24:35 -0700 From: Pavel Afonine To: PHENIX user mailing list Subject: Re: [phenixbb] how to fix a secondary structure? Message-ID: <[email protected]> Content-Type: text/plain; charset=UTF-8; format=flowed

Hi Xinghua,

you can use secondary structure restrains that are available in phenix.refine. It will automatically figure out secondary structure and apply proper restrains. This should preserve your helix geometry and still refine it.

Example:

phenix.refine model.pdb data.mtz main.secondary_structure_restraints=true

...

Recently I am trying to fix a helice, but found that it is so hard for me because of the poor resolution(3.2A) and also a unvisible loop near it, so is there any way to just fix a helice ? can I get it out of the whole structure to refine it?

If you still prefer not refining that helix over applying secondary-structure restrains, then you can do that too:

phenix.refine model.pdb data.mtz refine.sites.individual="not (chain A and resseq 123:156)"

where atom selection

chain A and resseq 123:156

selects your helix, and prefix "not" means you exclude it from refinement.

Of course you can do all this from PHENIX GUI.

Let me know if you have any questions.

Pavel.

------------------------------

Message: 4 Date: Fri, 16 Jul 2010 22:48:52 -0700 From: Pavel Afonine To: PHENIX user mailing list Subject: Re: [phenixbb] how to fix a secondary structure? - P.S. Message-ID: <[email protected]> Content-Type: text/plain; charset="utf-8"; Format="flowed"

By the way, more information about secondary structure restraints in phenix.refine you can find in recent PHENIX Newsletter (pages 12-17):

http://www.phenix-online.org/newsletter/CCN_2010_07.pdf

Pavel.

On 7/16/10 5:52 PM, xinghua qin wrote:

...

Hi phenixers: Recently I am trying to fix a helice, but found that it is so hard for me because of the poor resolution(3.2A) and also a unvisible loop near it, so is there any way to just fix a helice ? can I get it out of the whole structure to refine it?

Best regards

Xinghua Qin

-- Xinghua Qin College of Biological Sciences No.2, Yuan Ming Yuan West Road Haidian District?Beijing?China?100094 Tel: +86-10-62732672

------------------------------