The first step in performing a low-symmetry reconstruction is generating a preliminary model. This is accomplished by searching for particles with rotational symmetry and particles with mirror or near-mirror symmetry. These 2 views are then used to generate a very rough initial model. No CTF correction is done in this part of the procedure. The model generated at this stage is generally very poor. This is expected and is not a problem. Don't expect this model to look anything like your final reconstruction (except, perhaps for the view along the symmetric axis).
Determine defocus of particles from each micrograph. (ctfit)
Combine particles from micrographs with similar defocus and filter at the first zero. (proc2d)
Now you should decide if you want to do CTF correction or not:
Yes, I want to make a CTF corrected structure.
No, I'll try the uncorrected structure first.
Last Modified: