oversampling map

[Dominique]

Hello,

I'd like to oversample a 3dvolume prior to resampling it in chimera to another map and creating a composite map. Would decreasing my pixel size in eman2 be oversampling the map? What would be the command?

Thanks,

DG

[Muyuan Chen]

e2proc3d.py a.hdf b.hdf —process math.fft.resample:n=0.5

Should work I think. It will deal with pixel size automatically.

On Jun 14, 2022, at 1:18 PM, Dominique <domg...@gmail.com> wrote:

Hello,

I'd like to oversample a 3dvolume prior to resampling it in chimera to another map and creating a composite map. Would decreasing my pixel size in eman2 be oversampling the map? What would be the command?

Thanks,

DG

--
--
----------------------------------------------------------------------------------------------
You received this message because you are subscribed to the Google
Groups "EMAN2" group.
To post to this group, send email to em...@googlegroups.com
To unsubscribe from this group, send email to eman2+un...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/eman2

---
You received this message because you are subscribed to the Google Groups "EMAN2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eman2+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/eman2/9c67cdfc-cc4b-4b77-b215-359fcd6f42ean%40googlegroups.com.

[Steve Ludtke]

Worth pointing out that this depends on how you prefer to oversample. Muyuan's suggestion will perform Fourier oversampling with zero-padding. This may result in undesirable ringing in the interpolated voxels (depends on the structure). The alternative would be to use --scale=2 --clip=newbox which will perform real-space interpolation.

--------------------------------------------------------------------------------------

Steven Ludtke, Ph.D. <slu...@bcm.edu>                      Baylor College of Medicine

Charles C. Bell Jr., Professor of Structural Biology        Dept. of Biochemistry 

Deputy Director, Advanced Technical Cores                   and Molecular Biology

Academic Director, CryoEM Core

Co-Director CIBR Center

To view this discussion on the web visit https://groups.google.com/d/msgid/eman2/25EA3061-25CE-4B06-9FB9-CA7F164E0ABB%40gmail.com.

[Dominique]

I'd like to do Fourier space oversampling to minimize data loss. 

Attached is what I am seeing as displayed in Fiji. I am seeing these rings in the slices. Is this the undesirable ringing? When I converted the hdf to a mrc and the mrc opens up as 2d planes and not a 3d volume. I don't know why.

These are the commands I ran:

e2proc3d.py cryosparc_P2_J1115_003_volume_map.mrc cryosparc_P2_J1115_003_volume_map.hdf

e2proc3d.py cryosparc_P2_J1115_003_volume_map.hdf e2osfft_cryosparc_P2_J1115_003_volume_map.hdf --process math.fft.resample:n=0.5

e2proc3d.py e2osfft_cryosparc_P2_J1115_003_volume_map.hdf  e2osfft_cryosparc_P2_J1115_003_volume_map.mrc

I get a volume when I include --clip with a box size in-between the original and 2x the original box size. It seems unnecessary for the box to be 2x the size as its beyond the reconstruction but is there any particular reason for this? It seems necessary to give a box.

e2proc3d.py cryosparc_P2_J1115_003_volume_map.hdf e2osfft-clipbs600_cryosparc_P2_J1115_003_volume_map.hdf --process math.fft.resample:n=0.5 --clip 600,600,600

Is this clipping the box in real space after zero padding interpolation? If any ringing effect is getting clipped off I assume it's not an issue? 

[Steve Ludtke]

On Jun 15, 2022, at 3:31 PM, Dominique <domg...@gmail.com> wrote:

I'd like to do Fourier space oversampling to minimize data loss. 

Oversampling means you are making the images larger, ie - making a 256 x 256 image into a 512 x 512 image. There is no concern about data loss when oversampling. If you oversample in Fourier space, the values in between the original pixels may appear to have strange high frequency oscillations in them, depending on the behavior of the structure factor and filter applied to the original volume.

Interpolation in real space will result in non-zero values at high frequencies (though they will be small), but a smoother volume in real-space.

Attached is what I am seeing as displayed in Fiji. I am seeing these rings in the slices. Is this the undesirable ringing? When I converted the hdf to a mrc and the mrc opens up as 2d planes and not a 3d volume. I don't know why.

Does Fiji support 3-D volume display?  I've only used it a few times, and only for 2-D image processing. The MRC format does not really distinguish between stacks of 2-D images and single 3-D volumes. Technically the format has a way to specify the difference, but almost no software supports it. In EMAN2 if you use a .mrcs extension it will interpret volumes as stacks of 2-D images. If you use a .mrc extension it will interpret volumes as volumes. You can look at the results directly in EMAN2 irrespective of format. Just run 'e2display.py'. You can select 2-D slice-wise or 3-D visualization.

https://youtu.be/mTJ4gH7EOUA

https://youtu.be/tB3O5v_gYdg

These are the commands I ran:

e2proc3d.py cryosparc_P2_J1115_003_volume_map.mrc cryosparc_P2_J1115_003_volume_map.hdf

e2proc3d.py cryosparc_P2_J1115_003_volume_map.hdf e2osfft_cryosparc_P2_J1115_003_volume_map.hdf --process math.fft.resample:n=0.5

e2proc3d.py e2osfft_cryosparc_P2_J1115_003_volume_map.hdf  e2osfft_cryosparc_P2_J1115_003_volume_map.mrc

Ok, that's a bit silly. You can use whatever format you prefer for input and output.

e2proc3d.py cryosparc_P2_J1115_003_volume_map.mrc e2osfft_cryosparc_P2_J1115_003_volume_map.mrc --process math.fft.resample:n=0.5

should be fine. 

math.fft.resample automatically resizes the output. --scale does not, and must be preceded by --clip when making maps larger.

To view this discussion on the web visit https://groups.google.com/d/msgid/eman2/733503be-43b2-4841-94bd-7519a435d006n%40googlegroups.com.
<edgewithclip600.png><ringedges.png>