GSFSC unreliable in all refinements--due to preferred orientation?

hello,everyone!
I have a question about GSFSC curves. During data processing, Most of the curves show a similar situation, I tried to remove duplicate particles, But it’s still the same. I want to know what’s causing the fluctuations , And then what do I need to do about it?


Thanks a lot !

1 Like

Welcome @zzzzzz !

This is likely due to incorrect microscope parameters during import. I’d check again the microscope parameters specified when you imported micrographs.

You can read more about this in the guide (see Sharp bumps in FSC (CTF issues)) and in the RELION documentation.

Best,
Kookjoo

2 Likes

A few different options, but the easiest to check/test (after making sure you didn’t typo the Cs during import) is large beam tilt which hasn’t been corrected.

Try Global CTF refinement with just beam tilt (not trefoil) enabled, re-refine. Oscillations will probably go away.

However, you should also check pixel size carefully, as incorrect pixel size can also cause this effect.

It’s unlikely that the dataset is small enough and defocus uniform enough that you have holes in the signal.

1 Like

Thanks for you reply
I checked the microscope parameters and After re-uploading the data and performing the calculations, the same problem occurs, and other data did not present similar problems

Thanks for your reply.
I’ve tried Global CTF refinement, but But it’s the same as before.
I’ve also check pixel size, It matches the electron microscope I used.

And after several rounds of refinement , The Oscillations becoming more obvious.

What did the log of Global CTF look like - what did the plots look like for the aberrations and residuals?

Thanks again for your help, and is that right?



Is it possible that you accidentally imported with constant CTF on?


I’ve checked it again,“output constant CTF” is not on

Are there any other parameters that need special adjustment?

You have entered Cs as 27mm - should be 2.7mm I assume

5 Likes

And here we are!

(Always sanity check your input parameters…) :wink:

Thank you all for your patience and help, it was all my fault for being so careless, problem solved now!

2 Likes