IsoNet-cryoET / IsoNet

Self-supervised learning for isotropic cryoET reconstruction
https://www.nature.com/articles/s41467-022-33957-8
MIT License
67 stars 12 forks source link

Random half 1 and 2, Resolution Estimation #22

Open Zarrathustra opened 2 years ago

Zarrathustra commented 2 years ago

Thank you so much for this wonderful work!

One small question. I find that in your paper, the subtomos are randomly divided into two random halves. From each half, a predict is performed. Therefore, you will get two corrected tomograms, and make resolution estimation possible.

Screen Shot 2021-12-22 at 18 54 17

However, in your code, I can not find the corresponding part.

So, how can I generate two comparing corrected tomograms?

Looking forward to your reply.

Best wishes.

procyontao commented 2 years ago

Hi,

Splitting random halves is not in our code. What we did is to separate subtomo.star into two files using text editor, perform the refine step using each file.

Best wishes