Small soluble protein can't be well aligned in certain orientations

I think no-alignment classification in 2D is still only an option in Relion, sorry about the confusion. But the same idea works with alignment. And yes, this strategy may not work at all. Trying to think of ways to utilize all those particles while you gear up for next collection with hopefully improved grids…

@Pinf Did you make any progress with the data processing, or have you decided to try a different sample preparation or data collection strategy?

Hi,

I tried to 1) remove particles with lower resolution from the best data subset based on 2D classification (after 3D hetero-refinement) 2) apply mask on the stable region; 3) particle subtraction , but these tests don’t help.
I have tried to prepare samples with CHAPS, CHAPSO, FOM, β-OG, and Tween-20 (I didn’t got NP-40 then), but the samples are not good (particle density is pretty low for some of them, even I used almost 10-folds higher protein concentration). I collected ~200-300 micrographs for each sample. The best 2D result is shown below, and the views in 2D classes are only sides or top views with 2-balls shape. These views also appear in the previous data without detergent additives (shown above) and still can not be well aligned (noisy background). Maybe I need to prepare samples with very thin ice and collect data with tiled stage?

Thanks for asking!

1 Like