Ray / side-lobe density around ab-initio 3D volume despite clean-looking 2D major class

[rkyadav0889]

Hi everyone,

I am processing a protein in CryoSPARC and I am seeing a persistent artifact in my ab-initio 3D reconstruction. The main body of the volume looks reasonable, but there are many ray-like / side-lobe densities extending around the molecule. These side densities are problematic for interpretation and I am trying to understand whether this is a particle issue, preferred orientation / anisotropy, or something else.

I attached the 2D class averages and the corresponding 3D volume screenshot/map.

I re-extracted with a smaller box size to test whether nearby particles/fibrils were inside the box. The side rays still remained, so box size alone does not seem to solve the problem.

If the 2D classes look acceptable but 3D ab initio produces these rays, what would be the best next diagnostic step?

Has anyone seen similar ray-like density around a protein, and what solved it?

Any suggestions would be very helpful.

Thank you!

[olibclarke]

You have a very strong preferred orientation, which is leading to overfitting streaks in the other direction (and likely highly anisotropic density). I would suggest trying some high-CMC surfactant additives during grid preparation, as a starting point.

[rkyadav0889]

Hi Oliver,

Thanks for reply and suggestion.

In this sample we added 0.001% Tween-20 during grid preparation.

[olibclarke]

It may require some experimentation to find one that works for your sample - CHAPS, CHAPSO, fluor FC8, FOM, DM are all good candidates

[rkyadav0889]

Thank you Oliver, for helpful suggestions.