Hi,
After splitting exposures by image shift and refining improved parameters for beam tilt & trefoil, it would sometimes be useful to be able to then refine other parameters, such as anisotropic magnification, with a single group.
Currently, if we try to subsequently merge exposure groups and re-refine, this will throw an error, complaining that CTF params vary within a single exposure group.
Perhaps one way to address this would be to have an option in Global CTF refine in order to specify on a per parameter basis as to whether to refine per exposure group, or per dataset. This would be useful I think at least for anisomag, which I don’t think should vary between exposure groups.
Cheers
Oli
2 Likes
Thanks @olibclarke. We recorded this request.
1 Like
A slight modification of this might be useful also; e.g. for datasets which are quite heterogeneous (e.g. rotary ATPases), it would be good to be able to input multiple references/particle sets to Global CTF refinement (in a similar way to the Reference-based Motion Correction flexibility) so that beam-tilt parameters are uniform across the particle sets… or is there a way to copy beam-tilt/mag aniso/etc from one dataset to another (and appropriately adjusting the defocus/astigmatism values) that I’ve missed?
A case for this would be a postdoc in our lab currently, where per unique map there are not so many particles, which means there are even fewer in each exposure group, but combining each map and corresponding particle stack yields a good number of particles per exposure group. However, a unified refinement results in the extremely flexible complex being a poorly aligned mess at around 5-8 Ang, which does not yield good estimates of beam tilt etc.
2 Likes
Yes 100% co-signed! I think I’ve requested this before - there is probably a way using cs-tools but I am not sure…
Perhaps one way to make this easier would be to split out group/dataset-level CTF params as a different blob, so they could be swapped out using the low level inputs interface?
1 Like