Hello!
I just noticed a small inconsistency between duplicate removal and extraction that I thought I should mention (I hope this is the right category):
If particles from different branches are merged and they carry different alignments3D/psize_A values, remove_duplicate_particles can still successfully use alignments3D as the shift key when removing duplicates. However, extract_micrographs_multi with recentering enabled (recenter_using_shifts=true) appears to assert that all particles within a micrograph must have the same alignment pixel size, so those fail.
This is not a major issue because the particles can be split by alignments3D/psize_A before extraction, but I thought it was worth mentioning since the behavior differs between the two routines.
Cheers,
Sami