How the amount of frames does affect the accuracy of patch motion correction

Dear CryoSPARC community,

shall we expect that splitting of eer movies into 80 frames instead of 40 frames can increase the precision of the patch motion correction?

By the way, does the dose per frame make sense for eer movies? Or the dose per second is more important?

Thank you!

It’s a balance. You want enough signal per frame to permit accurate alignment/tracking of motion, but without excess motion within each frame.

I usually split EER frames so that dose rate is 1 e-/frame in CryoSPARC. Deviating much from that doesn’t make an appreciable difference.

Also note that RBMC can re-split the frame grouping so if you think you have excessive drift which can be corrected, you can always try later in the processing.

1 Like

Hi rbs_sci,
thank you for advice!