[phenixbb] Electron Density

Thanks, Nat, If the density level(map or coefficient) of phenix is not an absolute level, then I really do not need to worry what shows up by coot. I used the Fourier map generated by phenix, .mtz file. Thanks! 6. density level (CPMAS Chen)

7. Re: density level (Nathaniel Echols) 9. Re: density level (Pavel Afonine)

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

Message: 7 Date: Sun, 13 Jan 2013 12:21:00 -0800 From: Nathaniel Echols To: PHENIX user mailing list Subject: Re: [phenixbb] density level Message-ID: Content-Type: text/plain; charset=ISO-8859-1

On Sun, Jan 13, 2013 at 11:38 AM, CPMAS Chen wrote:

...

I recently updated my old phenix-1.6.4-486 to phenix-dev-1266. To my surprise, the absolute electron density on the 2Fo-Fc map is quite different.

The electron density maps (or coefficients) generated by Phenix are not on an absolute scale, and the values displayed by Coot are more or less irrelevant. The scaling method changed significantly between 1.6.4 and 1.8, so I'm not surprised that the map levels did too.

...

By the way, is there a simple way to keep only F and SIGF column of the autoxds generated truncate.mtz via phenix? the GUI does not allow to reduce columns. I normally use IMEAN and SIGIMEAN to do refinement, but this may not be necessary as it is pointed in the maillist, since the intensity is already transformed to amplitude after truncation in autoxds.

I can specify labels by xray_data.labels = "F, SIGF", but apparently phenix treats it as containing anomalous data. the following is extracted from the refine pdb file

MIN(FOBS/SIGMA_FOBS) : 1.16 REMARK 3 COMPLETENESS FOR RANGE (%) : 97.11 REMARK 3 NUMBER OF REFLECTIONS : 143207 REMARK 3 NUMBER OF REFLECTIONS (NON-ANOMALOUS) : 73247

Also, in the auto generated def file, xray_data.labels = "F, SIGF, DANO, SIGDANO"

This is intentional - why would you want to throw out information? However, you can always use either the reflection file editor (in the latest builds) or the French & Wilson GUI to convert IMEAN to F.

-Nat

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

Message: 9 Date: Sun, 13 Jan 2013 21:19:17 -0800 From: Pavel Afonine To: PHENIX user mailing list Subject: Re: [phenixbb] density level Message-ID: <[email protected]> Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"

Hi,

what kind of file you use to see the map in Coot: Fourier map coefficients (MTZ file) or actual map (CCP4 binary formatted file with the map)? I need to know this before offering a first guess on what may be the reason for this.

Pavel

On 1/13/13 11:38 AM, CPMAS Chen wrote:

...

Dear PhenixBBer,

I recently updated my old phenix-1.6.4-486 to phenix-dev-1266. To my surprise, the absolute electron density on the 2Fo-Fc map is quite different.

https://www.dropbox.com/s/prs0z8ayb73kg9k/Screenshot_sigma.png

At the same sigma level(1.5), the absolute density of the violet map is 0.32e/A^3 (phenix-dev-1266), while that of the blue map is 0.01e/A^3(phenix-1.6.4-486). Besides the different phenix used, there is no other difference on the data and model used.

Sam encountered a similar problem on the density.

http://phenix-online.org/pipermail/phenixbb/2012-December/019226.html

By the way, is there a simple way to keep only F and SIGF column of the autoxds generated truncate.mtz via phenix? the GUI does not allow to reduce columns. I normally use IMEAN and SIGIMEAN to do refinement, but this may not be necessary as it is pointed in the maillist, since the intensity is already transformed to amplitude after truncation in autoxds.

I can specify labels by xray_data.labels = "F, SIGF", but apparently phenix treats it as containing anomalous data. the following is extracted from the refine pdb file

MIN(FOBS/SIGMA_FOBS) : 1.16 REMARK 3 COMPLETENESS FOR RANGE (%) : 97.11 REMARK 3 NUMBER OF REFLECTIONS : 143207 REMARK 3 NUMBER OF REFLECTIONS (NON-ANOMALOUS) : 73247

Also, in the auto generated def file, xray_data.labels = "F, SIGF, DANO, SIGDANO"

Or there is a work round for this issue?

Thanks!

--

***************************************************

Charles Chen

Research Associate

University of Pittsburgh School of Medicine

Department of Anesthesiology

******************************************************

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