Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Confused about the simulation process #4

Open
eugenejyuan opened this issue Aug 29, 2022 · 0 comments
Open

Confused about the simulation process #4

eugenejyuan opened this issue Aug 29, 2022 · 0 comments

Comments

@eugenejyuan
Copy link

Hi, I have some confusion about the synthetic data generation process.

One is that I noticed the value of the original density volume data (.mrc) is processed by the following formula src/dataio.py#L177
$$mrc\_data = 2e4 * (power\_signal / power\_init) * (mrc\_data.shape[0] / projection\_sz[0]) * mrc\_data$$

  • the term $power\_signal$ is commented as the sum of squares in src/reconstruct/main.py#L162, but $power\_init$ is the sum of absolute values of the src data. Is this the expected behavior?
  • I didn't get a mathematical intuition of the process, why a 2e4 multiplier, why the term $mrc\_data.shape[0] / projection\_sz[0]$ not $projection\_sz[0] / mrc\_data.shape[0]$. Could you kindly give me some comments?

Another question is about the snr parameter. I see the default value is 10 in configfile configfiles/mrc2star_80S_128.ini#L5. Usually SNR in cryoEM is believed near 0.1-0.01 or even lower. In the code, the sigma of additive Gaussian noise is controlled by $power\_signal/snr$ src/noise_utils.py#L10.

  • Any guidelines about how to choose the two hyper-params? (Is there a more explicit relation between the hyper-param $snr$ and the actual physical SNR?) Is 0.1 (or even smaller number) a more appropriate value for snr?
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant