Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Requirement for --solvent_mask and --solvent_correct_fsc in relion_refine? #8

Open
olibclarke opened this issue Apr 18, 2024 · 3 comments

Comments

@olibclarke
Copy link

olibclarke commented Apr 18, 2024

Hi, in the misalignment correction section of the tutorial, it states:

image

Why are the latter two arguments needed? I am guessing for calculating the 3D-FSC? We usually only use a solvent mask in relion once we get to local angular searches, as using it during global search can lead to overfitting near the mask edges - but in this case, from what I glean from the preprint, it seems like we will want global search, as the alignment errors for anisotropic data can be quite dramatic. What is the best way to proceed?

As a sidenote re the manuscript, it might be informative to see quantification of per particle alignment errors with/without misalignment correction on, for a case where you have the ground truth alignments.

@procyontao
Copy link
Collaborator

You are right, mask may not be crucial here. There are two minor reasons: 1. 3DFSC need a mask (maybe for more accurate estimation?). 2. spIsoNet currently recognizes xx_unfil.mrc, which only generated with solvent corrected fsc. I think what you can do currently is to use a really big and smooth mask.

Quantifying per particle alignment error is a good suggestion, I will see what we can find.

@donghuachensu
Copy link

donghuachensu commented May 6, 2024

Hi, I remember I often do not use --solvent_correct_fsc in regular Relion auto-refine and it still generates xx_unfil.mrc files. So I don't fully understand your above statement "spIsoNet currently recognizes xx_unfil.mrc, which only generated with solvent corrected fsc". Does it mean that spIsoNet's xx_unfil.mrc files are only generated with --solvent_correct_fsc? Apparently Relion itself does not depend on --solvent_correct_fsc to generate xx_unfil.mrc files (please correct me if wrong).

@procyontao
Copy link
Collaborator

Hi,

I remember that RELION typically generates the xx_unfil.mrc file after the final iteration. But with the --solvent_correct_fsc flag, it generates the xx_unfil.mrc for each iteration, which is used by spIsoNet. Does that make sense to you?

Hi, I remember I often do not use --solvent_correct_fsc in regular Relion auto-refine and it still generates xx_unfil.mrc files. So I don't fully understand your above statement "spIsoNet currently recognizes xx_unfil.mrc, which only generated with solvent corrected fsc". Does it mean that spIsoNet's xx_unfil.mrc files are only generated with --solvent_correct_fsc? Apparently Relion itself does not depend on --solvent_correct_fsc to generate xx_unfil.mrc files (please correct me if wrong).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants