goddard at sonic.net
Mon May 4 10:24:02 PDT 2015
Sounds like the IMOD segmentation is using units of nanometers while the MRC map uses Angstroms. I don’t know if the IMOD segmentation tool uses the mrc file origin — if not, that could explain why you need to use origin 0,0,0 in Chimera to align. Does this IMOD segmentation file and MRC map file align when opened with IMOD? When you open the segmentation in Chimera (*.imod file) it will report the pixel size and scale that it read from the IMOD header in the Chimera Reply Log, e.g.
IMOD pixel size = 22.55 scale (1.0, 1.0, 1.0)
ctl.imod mesh opened
Chimera has no way to reduce the level of detail in an IMOD mesh to make the graphics go faster. It usually takes more than 10 million triangles to slow down a modern graphics card, and that would take an IMOD file at least a few hundred Mbytes in size. Is your file that large?
You can split a surface so that each of its connected components can be individually selected using
sop split #0
where #0 is the surface model you want to split.
But after you split it Chimera no longer knows that all the microtubules belong to one group. Also this sop split creates a new split model and hides the original surfaces. Reshowing the original surfaces requires “sop show #0”.
> On May 1, 2015, at 2:28 PM, Dougherty, Matthew T wrote:
> I have recently been getting some imod segmentation files and have been noticing three problems
> 1) when I read the MRC and imod files I am having a scaling problem, and the datasets do not integrate.
> For example, if the MRC voxel size is 31.22 31.22 31.22, and the origin index is 0 -114 0, I have to change the the voxel size to 3.122 3.122 3.122 , and the origin index to 0 0 0 so that graphics match correctly. Since there are number of people in the data chain before it gets to me, I thought it might check to determine if this is known problem before going upstream to unknown people to figure out if they are getting the imod output settings correct.
> 2) The graphics performance with imod segmentation files is sluggish in chimera, not sure what it is in imod. I assume that the format defines the vertices at the highest resolution corresponding with the MRC file. Is there a way to improve the performance on the chimera side? I am not using contouring, just reading the file in default mode.
> 3) The segmentation has a lot of components, such as vesicles and microtubules, that have different colors. Is there a way to break apart the components so they can be manipulated seperately? E.g., display/hide, change colors, translate and rotate.
> Matthew Dougherty
> National Center for Macromolecular Imaging
> Baylor College of Medicine
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu <mailto:Chimera-users at cgl.ucsf.edu>
> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users <http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users>
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Chimera-users