further polishing of unblur particles


I was wondering, based on a comment in your paper, about whether when using
the unblur, dose corrected ptcls in Relion for refinement, that one could
in fact also polish these particles successfully using the steps in Relion with
the original movies.

It seemed from your paper that you were/are developing your own approach to
the per particle tracking problem.


C Akey

Hi Chris,

You should absolutely be able to use the dose corrected particles for general Relion refinement, however i'm not sure it makes sense to use them for Relion's "particle polishing", as the weighted frames can have some pretty crazy filters on them, so Relion's b-factor estimation would probably fail. Having said that, you can write out aligned frames from Unblur, and if you apply the exposure filter, they will be written out with the individual filters on them if you wanted to play with it. You can also write out aligned and unfiltered frames from Unblur to use as input for the particle polishing if you wanted to go down that route. In the paper, the DLPs were aligned individually, basically just by cutting them out from the frames and running them through Unblur. For most of our current projects, we don't do individual particle alignment, we find the biggest gain comes from applying the exposure filter - however it is something we will likely investigate further in the future.