Patch CTF Fail right after start

CTF Estimation
1

Hi everyone,

We are encountering issues during the initial steps of data processing, and I’m unsure where else to look for answers. I apologize if my question sounds repetitive, but I have tried all the suggestions I found in the discussion forums so far.

Our Patch CTF estimation job keeps breaking approximately 20 seconds after it starts. The error message displayed is:

Traceback (most recent call last):

  • File “cryosparc_master/cryosparc_compute/run.py”, line 96, in cryosparc_compute.run.main*
  • File “cryosparc_master/cryosparc_compute/jobs/ctf_estimation/run.py”, line 373, in cryosparc_compute.jobs.ctf_estimation.run.run*
  • File “/home/cryosparc/software/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.8/site-packages/matplotlib/pyplot.py”, line 1755, in ylim*
  • ret = ax.set_ylim(*args, *kwargs)
  • File “/home/cryosparc/software/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.8/site-packages/matplotlib/axes/_base.py”, line 4027, in set_ylim*
  • bottom = self._validate_converted_limits(bottom, self.convert_yunits)*
  • File “/home/cryosparc/software/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.8/site-packages/matplotlib/axes/_base.py”, line 3614, in _validate_converted_limits*
  • raise ValueError(“Axis limits cannot be NaN or Inf”)*
    ValueError: Axis limits cannot be NaN or Inf

Initially, we suspected corrupted files and repeated the alignment job, but it didn’t resolve the issue. Next, we considered the possibility of corrupted files during data transfer. We identified the last FoilHole file that CryoSPARC could not process and removed the folder from the dataset, but this also did not work. We attempted to run the job with 1 or 2 GPUs, but it failed again. Surprisingly, when we ran the job with a random set of 1000 movies from our dataset (7,900 movies total), it worked! However, this amount of data is insufficient for our needs.

Subsequently, we attempted to process 3,900 movies (approximately half of our dataset) in the hopes that it would work similarly to the successful run with 1000 movies, but unfortunately, it failed again.

I’m afraid I’ve run out of ideas regarding what could be causing the job to break. It consistently runs about 6 micrographs before failing.

Does anyone have any idea what might be causing this issue?

Many thanks!

2

Hi @hciol

We have found that this is likely due to a corrupted movie file. Please review this post and let me know if it solves your issue.

Best,
Kye

1 Like
3

Hi, Kye

Thank you so much for your reply. It did work! There was a set of ~ 350 micrographs all black, and after excluding them, the CTF patch worked. So far, I had only been able to move on doing the full-frame alignment.

It didn’t occure to me to curate the exposures right after motion corr. Thanks again for sharing the post and the tip.

Bests,

1 Like