[Chimera-users] Place Objects output to mrc

[Ricardo Righetto]

Hi,

Regarding
https://www.rbvi.ucsf.edu/pipermail/chimera-users/2022-March/018516.html

I stumbled upon this thread somewhat late, so here is my approach to do
what Pranav asked, just in case:

1. Use the "Place Objects" plugin to map back your particles into the
tomogram, using STL files to represent the average

2. Generate an MRC volume made up entirely of "ones", exactly the same size
as your tomogram. For example, for a volume that is 928x928x464 voxels one
can do this in EMAN using:
$ e2proc3d.py :928:928:464:1 ones_bin4.mrc
You could obviously use some other program to do the same thing, or Python,
MATLAB, etc.

3. Load this "ones" volume in your Chimera session. Make sure all regions
are loaded in step 1, e.g.
vol #2 style surface step 1 level 0.5 region all
(the volume doesn't have to be visible, though)

4. Mask your "ones" volume using the surface of the mapbacks from "Place
Object", e.g.:
mask #1 #2 fullmap true

5. Save the new, masked volume as an MRC file

The result is essentially a binary volume (well, not exactly because of the
masking interpolation) where voxels of your mapped-back particles are 1 and
the rest is 0. This is quite useful to compose segmentations and mapbacks
on tomograms in Chimera or ChimeraX having full control like we do with any
MRC volume (unlike e.g. STL or COLLADA objects).

Note you may get some artifacts on the edges, so you might have to play a
bit with the thresholds both when creating the STL of your subtomogram
average and when displaying your new volume "sculpted" by the mapbacks, to
minimize those. The results however have been good enough in my experience,
working with low-resolution averages :-)

Hope this helps!

Best wishes,

--
Ricardo Diogo Righetto
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.rbvi.ucsf.edu/pipermail/chimera-users/attachments/20220520/8e7738ec/attachment.html>