How to extract a 3D density volume slice in reciprocal space

Cryo-EM Data Processing
1

Hi, community!

I am implementing a Python script to quantify the agreement between an experimental Cryo-EM particle image and a re-projection of the corresponding 3-D reconstruction.
The pipeline is as follows:

  1. Reconstruct a consensus 3-D volume M in cryoSPARC.

  2. Compute the 3-D Fourier transform of M (without normalization) to obtain Mʹ.

  3. Center the zero-frequency component of Mʹ.

  4. Extract a central 2-D slice S from Mʹ at the pose P reported for the particle.

  5. Transform the experimental particle image I (corresponding to pose P) into reciprocal space and apply CTF correction to yield Iʹ

  6. Center the zero-frequency component of Iʹ.

  7. Evaluate the correlation between S and Iʹ.

Because S and Iʹ should share the same pose P, their correlation is expected to be high. In practice, however, the value is essentially zero, indicating no detectable correlation.

Are there any steps I might be missing in my current pipeline?