Hi CCPEM,
Consider a large particles.star file with "old" (refined) and "new", freshly picked particles. The "old" particles went through "3D auto-refine" already and have (nearly) optimal parameters (shifts and angles). The reconstruction of the "old" particles alone already results in a high resolution map (e.g. 5A). Such a situation arises during on-the-fly data processing during data collection.
Is it now possible to tell "3D auto-refine" which particles to use for alignment (E-step)? Ideally on a per-particles basis in the star-file. That way it would be possible to only align "new" particles (E-step) while using all particles for the the reconstruction at the end of each iteration (M-step).
By doing so I wouldn't need to align all particles again and again, when adding new particles to the star-file.
I am aware that such an intervention likely introduces artefacts or non-optimal parameters. However, that operation would only be applied in the beginning followed by more thorough refinement once data collection has finished.
Would are there any direct or indirect per-particle rln-Parameters that could be used for that? Is it possible to use the external_reconstruction function to reconstruct all particles while only aligning the "new" ones?
Thanks
Ferdinand
########################################################################
To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCPEM&A=1
This message was issued to members of www.jiscmail.ac.uk/CCPEM, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|