Re: [phenixbb] Waters in FEM

Could you please try latest Phenix from nightly builds or send me files off-list and I will try right now? Pavel On 10/22/14 3:22 PM, Murpholino Peligro wrote:

I also can see something like that... https://www.dropbox.com/s/lfd2sz44b4yzu5c/water19.png?dl=0

purple is FEM and blue is normal map.

2014-09-09 1:36 GMT-05:00 Pavel Afonine mailto:[email protected]>:

Hi John,

thanks for feedback, this is very valuable as I'm still gaining experience with FEM and tuning the protocol based on it.

I can't tell what exactly happens unless I have data and model files and can reproduce the map you get. Would mind sharing the files off list so I can have a closer look?

Thanks! Pavel

On 9/7/14 2:38 PM, Tanner, John J. wrote:

...

Well, this is embarrassing...the assign label window was hiding behind the graphics window.

Now that I do see the FEM, I have an observation about the FEM. Although the FEM is providing an informative assessment of the protein and bound ligands, the assessment of water is a little strange. There are several water molecules that have strong 2Fo-Fc density but absolutely zero FEM density. Even if I contour the FEM down to the lowest possible value (0.09 sigma in this case), I still see no FEM density on these waters. The waters in question have good hydrogen bonding to the protein and acceptable B-factors (26-30 Ang^2). These are waters that I would normally keep in my final model. Has anyone else seen this behavior?

Some examples are posted here:

http://faculty.missouri.edu/~tannerjj/FEMwaters/FEMwaters.html http://faculty.missouri.edu/%7Etannerjj/FEMwaters/FEMwaters.html

On Sep 7, 2014, at 2:51 PM, Pavel Afonine wrote:

...

Hi John,

it does not open via Auto Open MTZ.

start Coot then

File -> Open MTZ, mmCIF, fcf or phs -> the choose FEM,PHIFEM (there are two Fourier maps in this file : FEM and usual 2mFo-DFc).

Pavel

On 9/7/14 11:14 AM, Tanner, John J. wrote:

...

Has anyone had problems reading feature enhanced map mtz files into coot? I made an FEM on a linux computer with phenix.fem, which created an mtz file:

-rw-r--r-- 1 tannerjj tanner 10760960 Sep 7 12:21 fem.mtz

As you can see, the mtz file is not empty. The fem job appeared to have ended normally, since the last line in the out file is

FEM loop: done so far: 100%Time: 3112.3967

However, when I read this mtz into coot, the map does not appear. There is no error message from coot. I tried autoinput mtz and read mtz. I'm using coot on a mac. I have not tested with coot on Linux.

Thanks.

Jack Tanner

John J. Tanner, PhD Professor of Biochemistry and Director of Graduate Admissions and Recruitment Professor of Chemistry (Joint Appointment) University of Missouri-Columbia 125 Chemistry Building Columbia, MO 65211 email: [email protected] mailto:[email protected] phone: 573-884-1280 fax: 573-882-2754 http://faculty.missouri.edu/~tannerjj/tannergroup/tanner.html http://faculty.missouri.edu/%7Etannerjj/tannergroup/tanner.html

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

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