Hi, all!
I’m currently stuck in a problem when dealing with bin1 particles.
I imported .mrcs form movies with raw pix size of about 0.5 and then did whole frame motion correction and CTF estimation. After rounds of classification, I extracted bin1 particles with box size of 512 pix. But when I performed ab initio recontruction, the reconstruction box size is always 128 or 64. I don’t know why this happened.
Generally, I use Motioncor2 for motion correction, and pix for imported micrograph is about 1. So I use 256 pix for particle extraction and things went on well when ab initio is done after such processes.
Ab-initio downsamples depending on the max resolution it is set to use, so it is perfectly normal that your ab-initio results are a smaller size than your input particles. If you continue with refinement, the full resolution of the dataset is used.
In my previous experiment, downsampling of particles never happened by default during ab initio. I’m wondering if this is due to update or if it is because of the large box size.