Merging two data sets with different pixel size

open

#1

Dear,
Can we merge two data sets with different pixel sizes on the latest Cryosparc? Could you please explain any way to do that?


#2

Hi @ananthkoch,

Currently, this is difficult to do. On-the-fly magnification resampling is on the way that will make this easier, however, it is very rare that you would actually have two datasets with different pixel sizes where merging them is actually useful. Generally speaking, if the datasets were collected on different microscopes or different cameras or different samples (ice thickness etc) then one of the datasets will have (significantly) more signal at high resolutions than the other dataset and therefore will dominate during refinement, leaving the second dataset to contribute relatively little to the final result.
In our test so far we have not found a good example of two datasets that can be merged yielding an improved result, but our testing has not been extensive since this is not yet a common practice.

A simple/ad-hoc way to combine two refinements is to simply resample the final maps on to the same pixel size after refinement and then average the maps. This can be done with eg. Chimera.


#3

Hi,
The relion3.1 can merge two different pixel size data sets. maybe useful?


#4

Thank you for the answers. still I expect Cryosparc will develop way to merge data sets.


#5

Hi,

Could you please explain more how to do so?


#6

Hi,
you can see it at the relion website.