Le 2018-09-17 15:52, Dmitry Lyumkis a écrit :

Dear Stefano, 

One approach that we like is to simply calculate the cross-correlation coefficient of the map to the model (converted to a map) in Chimera (options “Fit in map” and "use simulated map from atoms”). Then adjust the voxel size in the “coordinates” tab for the cryo-EM map, and perform essentially the analysis that you’re doing. In my experience, this tends to provide a sharper peak, but you may have to perform for the asymmetric unit of the model. 

Best, 

Dmitry

 

 

On Sep 17, 2018, at 12:19 PM, Sjors Scheres <[log in to unmask]> wrote:

Dear Stefano,

At this resolution, re-refining your model may not yield a good absolute
scale, as you have noticed. If you have a homologous X-ray structure for
the virus, then you could try rigid-body fitting that into the map at
different scales and see which one fits best (so without re-refinement).
Best would be to have an independent measurement of the pixel size on
the same scope, for example by solving a known structure. Perhaps some
of your colleagues at the institute have determined the pixel size at
this magnification already?

HTH,
Sjors



On 09/17/2018 11:08 AM, Stefano Trapani wrote:

Dear all

We are refining the atomic model of an icosahedral capsid based on a
cryoEM map obtained from samples of intact viruses. The nominal pixel
size of the map is 1.21 A and the resolution (according to Relion) is
3.9 A.

We have tried to finely calibrate the map pixel size by running
several refinements of the atomic model at different map magnification
levels (phenix.auto_sharpen + phenix.real_space_refine with 2ry
structure restraints and no ADP refinement).

Results are quite confusing (see attached graphs, the nominal pixel
size is represented by the black thick vertical line):

 * the best map-to-model fit is found at a rather larger pixel size
   (~1.35A) than the nominal one (1.21 A);
 * however the resulting icosahedral assembly is quite loose around
   1.35 A pixel size (according to PISA predictions the 60-meric
   would not be stable);
 * also, quite a few Ramachandran conformations initially in favoured
   regions shift towards less favoured (though allowed) regions as
   the map-to-model fit increases.

Do you have any suggestion on how to interpret these results or a
better procedure (if any) to better evaluate the map magnification ?

Cordially

-- 
Stefano Trapani

Maître de Conférences
http://www.cbs.cnrs.fr/index.php/fr/personnel?PERS=Stefano%20Trapani
-------------------------------------
Centre de Biochimie Structurale (CBS)
29 rue de Navacelles
34090 MONTPELLIER Cedex, France

Tel : +33 (0)4 67 41 77 29
Fax : +33 (0)4 67 41 79 13
-------------------------------------
Université de Montpellier
CNRS UMR 5048
INSERM UMR 1054
-------------------------------------

-- 
This message has been scanned for viruses and
dangerous content by *MailScanner* <http://www.mailscanner.info/>, and is
believed to be clean.
------------------------------------------------------------------------

To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1

-- 
Sjors Scheres
MRC Laboratory of Molecular Biology
Francis Crick Avenue, Cambridge Biomedical Campus
Cambridge CB2 0QH, U.K.
tel: +44 (0)1223 267061
http://www2.mrc-lmb.cam.ac.uk/groups/scheres

########################################################################

To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1

--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

---

To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1

To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1