[Chimera-users] MRC map values -128 to 127

Bernard Heymann bernard_heymann at nih.gov
Fri Sep 7 13:47:38 PDT 2007


Tom

The key problem is that people keep changing the "standard" for the  
MRC format. Here is the current official version:
http://www2.mrc-lmb.cam.ac.uk/image2000.html

So it is not so simple any more, as the software from the MRC and  
associated packages now probably adhere to the new standard, while  
other packages and old MRC format files still use the unsigned 8bit  
form. The suffix ".mrc" is used for both, so that cannot be used as a  
distinguishing mechanism. Also, there are significant differences  
between the CCP4 (suffix ".map") and MRC formats, so you cannot  
substitute the one for the other.

I'm still figuring out how to handle it in Bsoft.

On Sep 7, 2007, at 4:11 PM, Tom Goddard wrote:

> Hi Bernard,
>
>  Thanks for the info on the history of signed versus unsigned 8-bit  
> MRC files.  I see that CCP4 clearly uses signed 8-bit and most MRC  
> used for EM data uses unsigned.  Is there any definitive  
> specification for MRC format?  I just made up that bit about an  
> "MRC standard".  I'm not aware of anyone publishing a standard.   
> The Chimera change was motivate by the EM Databank using CCP4 for  
> their maps.  A possible more explicit handling as you suggest is to  
> assume signed 8-bit for file suffixes ".ccp4" and ".map" (used by  
> EMDB) and unsigned 8-bit for the file suffix ".mrc" -- basically  
> distinguish CCP4 from MRC format in Chimera.  I'm not sure if the  
> ".map" suffix used by EMDB to mean CCP4 is in use by other EM labs  
> to mean MRC.
>
>    Tom

Bernard Heymann, Research Fellow
Rm 1515, 50 South Dr., MSC 8025, NIAMS, NIH
Bethesda MD 20892-8025
Tel. 301-451-8241, Fax. 301-480-7629



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimera-users/attachments/20070907/a0115578/attachment.html>


More information about the Chimera-users mailing list