I want to study the effect of dose on my particle reconstruction and wanted advice on what would be the best way to go about it.
Since I have EER fractions, if I would just remove the last few frames during motion correction would that mean that my total electron exposure would also be reduced? (and in this case would still doing dose-weighting make sense?).
For context, the data was collected in an energy filtered Krios with input dose of 45 e/A2.
Yes, some people remove - or used to - the first few and last few frames, no matter which file format you use. The motion correction plots help deciding which frames to remove, if any. Removing the first ones has nothing to do with dose, but with motion.
(the notion of “frames” is aritificial in EER, by the way)
Dose weighting always makes sense IMO, what you can try to do is to tweak the default parameters to see if you can improve the current reconstructions. But to be honest it seems to me you are already at a relatively low dose. You might sacrifice too much signal if you go lower.
There are Chris Russo’s papers - and certainly others - about this.
All that said, I am sure I am not the best placed to answer this, though, so looking forward to reading other people’s opinions.
In contrast to the first few frames having the most motion (which they very often do) which makes them tempting to axe, they also contain your untainted minimally exposed versions of particles. for this reason if I wanted to “chop” dose off then I would chop from the end and pretend you had collected less time.
Yes, thanks for the link. I mentioned ‘or used to’ because I haven’t processed data from scratch recently, maybe the local motion correction is dealing nicely with the initial motions. Of course one should keep those images as much as possible because they also contain higher resolution information.