Hi all,
Recently, I’ve been reworking some old data sets to try to reduce anisotropy in a couple of my maps. In this process, I’ve been screening particles pretty rigorously, trying to select the best 2D classes only and obtain relatively equal representation of each class. I typically will extract particles first with a 4x bin, then I will unbin them after refinement yields a decent structure, refilter with 2D and 3D classification, and run a final refinement job. This typically works fairly well.
However, this week, I’ve been encountering the same issue across all my data sets after obtaining a decent structure at ~4 A using 4x binned particles. Upon unbinning and reextracting my particles, rescreening them, and running either homogeneous or NU-refinement, my FSC curves no longer cross zero and the relative frequency component of the cFSC plot shows most or all of my particles in a single bin at the highest resolution (plots attached).
I’ve removed duplicate particles, changing the distance to correspond to the particle size, and again, this is happening when my particles are extracted without binning, so the resolution should not be limited in that way. I’ve used this method in the past with no issues on these data sets. Does anyone have any suggestions for what might be going on?
Thanks!
Kate