[3dem] [ccpem] particle picking in crowded micrographs

[Marin van Heel]

Dear Bastiaan and All here interested:


A more subtle two-stage particle picking procedure for highly concentrated
samples is implemented in IMAGIC which is illustrated in:

Afanasyev P, Seer-Linnemayr C, Ravelli RBG, Matadeen R, De Carlo S,
Alewijnse B, Portugal RV, Pannu NS, Schatz M, Van Heel M (2017).
Single-particle cryo-EM using alignment by classification (ABC): the
structure of Lumbricus terrestris haemoglobin. *IUCrJ.* *4*, 678–694.   (free
download)


[image: image.png]

Cheers

Marin van Heel

On Mon, Nov 27, 2023 at 3:32 PM Bastiaan Bergman <
bastiaan at modelmedicines.com> wrote:

> All,
>
> I'm reprocessing this data from Empiar
> <https://urldefense.com/v3/__https://www.ebi.ac.uk/empiar/EMPIAR-11176/__;!!Mih3wA!CUyE02u7zsgfaew-RbTyNrOq6oGiT2gUKjW0bIRXWVGHTDfNrRK_22uSkfU72hnZOQFy1aEZ2ZvBgA9YfmsgNNaMbV_bow$ > with Relion (4.0.1 commit:
> 8e4597), clearly it's possible to get a highres structure with it, they
> <https://urldefense.com/v3/__https://journals.plos.org/plospathogens/article?id=10.1371*journal.ppat.1011206__;Lw!!Mih3wA!CUyE02u7zsgfaew-RbTyNrOq6oGiT2gUKjW0bIRXWVGHTDfNrRK_22uSkfU72hnZOQFy1aEZ2ZvBgA9YfmsgNNZePmjmjw$ >
> did it with Cryospark.
>
> I'm having trouble picking particles. I tried autopicking with Topaz and I
> tried manual picking. The first did not result in any good particles (as
> judged by the 2D classes). Looking at it manually it seems too crowded or
> aggregated to really pick individual particles.
>
> There are 7000 micrographs, I looked manually at about 700 (10%) of them,
> the one below is one of the best 10 I could find. I have trouble
> picking single particles in this, most structures here seem to be larger
> than a single particle.
> [image: Screenshot 2023-11-27 at 14.55.44.png]
>
> Are there any expert directions I can get? How do I pick particles, is
> there hope with this micrograph or should I use different displaying
> settings? (Above was done with Topaz denoising). Or are there any tools to
> go through all 7000 and look for some better ones?
>
> Thanks for all your insights! I attached the micrograph.mrc
> <https://urldefense.com/v3/__https://drive.google.com/file/d/1mbEepGxoCt2zCEw3mc1HfDt6r5MWYlJ_/view?usp=sharing__;!!Mih3wA!CUyE02u7zsgfaew-RbTyNrOq6oGiT2gUKjW0bIRXWVGHTDfNrRK_22uSkfU72hnZOQFy1aEZ2ZvBgA9YfmsgNNaqqo6XeQ$ >
> here as well.
> Bastiaan.
>
>
>
>
>
>
> ------------------------------
>
> To unsubscribe from the CCPEM list, click the following link:
> https://urldefense.com/v3/__https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCPEM&A=1__;!!Mih3wA!CUyE02u7zsgfaew-RbTyNrOq6oGiT2gUKjW0bIRXWVGHTDfNrRK_22uSkfU72hnZOQFy1aEZ2ZvBgA9YfmsgNNZ6rEq9SA$ 
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20231128/1ec2a942/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screenshot 2023-11-27 at 14.55.44.png
Type: image/png
Size: 2738520 bytes
Desc: not available
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20231128/1ec2a942/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 1051927 bytes
Desc: not available
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20231128/1ec2a942/attachment-0003.png>