Manually defined dose-weights?

Data Processing Strategy
1

Hi there,

I’m working on a project that involves solving multiple structures from the same movies/particles, but from different frames of each movie (one structure from the first 10 frames, another from the next 10 frames, etc.). I’m trying to figure out the best way to do this in CryoSPARC. I’ve already tried specifying start and end frames in motion correction to no avail (motion correction is impossible in the later frames). So, I have two questions:

  1. would it be possible to run motion correction using the full stack, and then only produce a corrected micrograph using the motion trajectories from certain frames? This would have the benefit of using the information from the excluded frames, leading to a more “homogenous” tracking quality across the various resulting micrographs. From what I can gather, this is not possible, but I’d like to see if anybody else knows how.

  2. my next idea would be to modify the dose-weights manually, simply by setting 0 for all other frames outside of the desired ones for that structure and 1 for the desired ones, at all spatial frequencies. I know this will lead to a lower-quality structure, but that is fine for the purposes of this project.

I plan to use some strategy to create the micrographs for each set (either one of the above, or something else), then use the particle/orientation information from the reconstruction done using a standard workflow to directly make reconstructions from each set of micrographs.

Thanks in advance!

-Shervin Nia

2

I was testing something similar with more exhaustive fractionation conditions some time ago. I found it a lot easier to carry out motion correction in RELION as Bayesian Polishing can use specific frames, and import the particle stack into CryoSPARC if you need NU Refine, 3D Variability Analysis, 3D Flex etc.

  1. Can be done with RELION Bayesian polishing (since custom frame ranges are not supported in RMBC and you can’t successfully motion correct later frames. You could also try whole frame motion correction if patch fails.

  2. Just disable dose weighting (or use the non-dose-weighted micrographs which CryoSPARC generates automatically)

3

@shervinnia, thanks for the feedback. We’ve recorded this as a feature request, as there isn’t a straightforward way to do it in CryoSPARC at the moment. Thanks rbs_sci for providing some possible workarounds.