[Chimera-users] Origin issue of mrc file from chimera to pymol

[Tom Goddard]

Hi Kai,

  MRC map files have two header fields for specifying the origin in two different ways.  This problem with the format comes from the fact that the original origin specification was only for x-ray maps and could only handle integer grid points as the origin.  For electron microscopy it is sometimes necessary to specify an origin that is between grid points, for example when a symmetry axis of the map does not pass through a grid point but it is desired to make the origin lie on the symmetry axis.  So the MRC 2000 format added a floating point origin to handle electron microscopy maps.  Chimera reads and writes both origin fields and has been extensively tested that it follows the conventions for these fields.  Other software doesn’t always handle the origin header values, sometimes just ignoring them.  I don’t know what PyMol does.  It is a clue that you say it works if you use the entire map but not if you have cropped the map.  When you crop it changes the origin fields.  Software that ignores the origin header values will then not align such a map correctly with the original full map.

  At any rate, I believe Chimera is handling the origin values correctly and I don’t know any tricks to make other software correctly respect the origin specified in the file.

  Glad to hear that your export to BRIX format gave proper alignment in PyMol.  That is probably the way to go.  The only drawback might be that BRIX has only 8-bit data values while MRC maps are most often 32-bit floating point values.  The detectors used to acquire EM data are typically 10-16 bits I think but there is a good bit of noise so 8-bits is probably ok.

	Tom


On Sep 4, 2014, at 5:15 AM, Kai Zhang  wrote:

> Dear Chimera users,
> 
> I found the pdb files fitted to mrc maps in Chimera do not always fit
> the same map opened by Pymol. It seems that the origins are calculated
> differently between Chimera and Pymol. I can easily handle an entire map
> of mrc file. But if the map has been cut, it's very difficult to find
> out what really needs to be modified to fit both chimera and pymol. Does
> anyone know an easy way to deal with this issue?
> 
> Best wishes
> Kai
> 
> 
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
>