Hi CCPEMers,
When generating signal subtracted projections in RELION, and assuming that the region of interest is large enough to be amenable to either a regular masked refinement or a refinement of signal subtracted particles, which orientations and model are the best to use for signal subtraction?
Should one use the particle orientations and _class001.mrc (with the region of interest masked) from a focused refinement on the region of interest? Or the orientations and model from a “vanilla” refinement of the entire particle?
I have tried both with good results but I haven’t systematically compared the two, and I’m curious as to which is the better option in theory.
Cheers,
Oli
|