missing wedge with subtomograms

[Dave M]

Dear Eman2 community,

I have a general query regarding removing the wedge in tilt-series specifically for subtomogram averaging. Sorry this is a naive query but it will help me understand better.

I will consider a very simple unusual case (image attached) assuming there are thousands of particles with a preferred orientation perpendicular to the ice/grid plane but with random in-plane orientations (0-360 degree). If I perform subtomogram averaging on all these particles would this help removing the wedge effect? If I understand correctly, there will be a predominant wedge problem, say, if I am looking at only one particle but if I construct a map from subtomogram averaging thousands of particles there should be a minimal wedge effect (even for the particles with preferred orientation shown in the image). Is this correct? 

Thank you.

best wishes,

Dave

[Muyuan Chen]

In this case, it sounds like your particles have essentially the same Euler angle (alt, az), just different in plane rotation (phi)? If so, averaging many particles will reduce the missing wedge to a missing cone, but won't fill the Fourier space completely.  

I am sure Steve has a long answer to this...

Muyuan

--
--
----------------------------------------------------------------------------------------------
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/CAL1266mPKTJXOyA2Qms8ZZRMb2Mh3B7NVpsL7u%2B8dYqDx9QDNQ%40mail.gmail.com.

[Ludtke, Steven J.]

If all of your particles are in the same orientation except for a random in the plane of the ice rotation, then after subtomogram averaging you will have a missing cone, rather than a missing wedge. This is analogous to the old single particle random conical tilt method.

---

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 Technology Cores                  and Molecular Pharmacology
Academic Director, CryoEM Core
Co-Director CIBR Center

--
--
----------------------------------------------------------------------------------------------
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/CAL1266mPKTJXOyA2Qms8ZZRMb2Mh3B7NVpsL7u%2B8dYqDx9QDNQ%40mail.gmail.com.

<image.png>

[Ludtke, Steven J.]

Apparently we both answered basically the same thing at the same time. Take a look at the old RCT literature for a better understanding of some of these issues. 

---

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 Technology Cores                  and Molecular Pharmacology
Academic Director, CryoEM Core
Co-Director CIBR Center

To view this discussion on the web visit https://groups.google.com/d/msgid/eman2/CAO_0xqF9-h%2BOXt%3DzA8YiE83qkGq8Rt9hhcrF_A1aPxO5Ab7pMA%40mail.gmail.com.

[Dave M]

Hi Muyuan and Steve,

Thank you. I missed to mention that particles can also have a 180 degree flip (upside down); not all of them are necessarily facing up as I showed in the image, say half are facing up and half are facing down.

So for these preferred orientations I believe it would never be possible to obtain a full map. I will appreciate it if you have any thoughts or reference pointers if people have tried to fix this problem to obtain sub 10Angstrom maps; (for the case of preferred orientation particles without changing sample prep conditions to obtain random orientations)

Thank you.

best wishes,

Dave

To view this discussion on the web visit https://groups.google.com/d/msgid/eman2/6C273F2B-1611-4518-8E94-E281BC897E24%40bcm.edu.

[Tapu Shaikh]

On 3/2/24 03:47, Dave M wrote:
> I missed to mention that particles can also have a 180 degree flip
> (upside down); not all of them are necessarily facing up as I showed
> in the image, say half are facing up and half are facing down.

For the purpose of filling orientation space, a 180-degree flip doesn't
make any difference.  As Steve noted, it's the same problem that one
faces in random conical tilt.

> I will appreciate it if you have any thoughts or reference pointers if
> people have tried to fix this problem to obtain sub 10Angstrom maps;
> (for the case of preferred orientation particles without changing
> sample prep conditions to obtain random orientations)

With your preferred orientation and assuming no symmetry, you're going
to have anisotropic resolution.  There are tricks for filling in some
missing data; projection onto convex sets is the one I know of, although
I never tried it with high-resolution data.  I don't know if the
missing-wedge deep-learning methods work on missing cones.

If you can change your sample-prep conditions to avoid a preferred
orientation, I'd suggest that.  Also, if you have a suspension in ice
and still have a preferred orientation, I'd worry that a likely cause
for a preferred orientation is the air-water interface, which may harm
your specimen and prevent high resolution.

Hope this helps.

-Tapu Shaikh

[Dave M]

Helo Tapu, Muyuan, Steve

I just saw this paper: https://www.nature.com/articles/nmeth.4347

where they use tilt angle SPA tested with proteins with preferred orientations. Do you think it will work for the unusual case I showed in the cartoon image?

Thanks.

best wishes,

Dave

--
--
----------------------------------------------------------------------------------------------
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/d6bce937-6fd5-9328-1f45-a49e512c7bb2%40gmail.com.

[Ludtke, Steven J.]

There has also been some followup work since that paper getting into more details, see papers authored by Phil Baldwin. You are already tilting when you do tomography. The idea of doing some tilting with single particle collection is nothing new. Again, refer to random conical tilt papers from the 90s - 2000s. The requirement for completeness is that you have a sufficient number of samples at all points in Fourier space. Each image samples a plane in Fourier space. The missing wedge is what you get when you tilt those planes along a single axis. A missing cone is what you get when you take a missing wedge and rotate randomly in-plane. 

The argument in that paper is that preferred orientations are rarely as constrained as you are describing. That is, there is a preferred orientation, but actual particles are found in a range of orientations near that single orientation, not just a single exact orientation. If the width of the distribution around that point is, say, 20 degrees, and you are able to collect tilt data up to, say 60 degrees, then your missing cone is only 10 degrees wide, which is a lot better than a missing cone which is 30 or 45 degrees wide. It still isn't ideal, but it's a lot better.

---

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 Technology Cores                  and Molecular Pharmacology
Academic Director, CryoEM Core
Co-Director CIBR Center

To view this discussion on the web visit https://groups.google.com/d/msgid/eman2/CAL1266n4DO0QySu3Ji39XqrZdhSB6D%3DqQ-9AQkLZKpJ7ubhfbg%40mail.gmail.com.

[Dave M]

Thank you very much Steve, as always, for an easy to understand explanation. Now I get the point. 

best wishes,

Dave

To view this discussion on the web visit https://groups.google.com/d/msgid/eman2/5DA426BD-CC8E-47CC-B3C5-1D9271EFC06C%40bcm.edu.