FSC is not dropping to 0 - 3D Refinement

Cryo-EM Data Processing
1

Hi everyone,

I’m running the final Non-Uniform Refinement in CryoSPARC, but I’m encountering an issue with the FSC curve.

Here are some details:

  • Final particle number: ~230,000

  • particle diameter : 100 Å

  • Refinement: Non-Uniform Refinement

  • Symmetry: C1 (or your symmetry if different)

  • I already performed one round of **Remove Duplicate Particles （**min_dist_A at 40 Å) before the final refinement.

  • The refinement finishes normally, but the FSC curve does not drop to 0 at high spatial frequencies. Instead, it remains above zero.

I’m wondering what could cause this behavior.

Could this indicate that:

  • there are still duplicate particles remaining?

  • overfitting or masking issues?

  • some other processing problem?

Has anyone experienced something similar? Any suggestions on what I should check would be greatly appreciated.

Thank you!

Screenshot 2026-06-25 at 12.05.25
Screenshot 2026-06-25 at 12.05.251192×1862 227 KB

2

Likely some duplicate particles. Have you tried remove duplicates?

1 Like
3

Yes , I do removing duplicate particles before refinement

4

Increase the distance and check again. I would suggest 60-80% of your particle diameter.

2 Likes
5

(post deleted by author)

6

Thanks very much for your advice. I then removed duplicate particles using min_dist_A = 80 Å (my protein is approximately 100 Å in size), but I still obtained similar results.

Screenshot 2026-06-26 at 10.57.28
Screenshot 2026-06-26 at 10.57.281226×1968 249 KB

7

What CS version are you using? In earlier versions there was an issue with remove duplicates:

What happens if you manually specify the micrograph pixel size?

Oli

8

Thanks, I’m using CryoSPARC v4.7.1. The particles were picked and extracted in v4.7.1, and no Fourier cropping or binning was used, so the particle and micrograph pixel sizes are identical.