Cryosparc live, patch ctf correction


I was wondering if there’s some issue that came up with recent 4.2 or 4.3 versions of cryosparc live, we had no issues running ctffind on our data before and now i have to check the ‘classic mode’ to use the old patch ctf algorithm before version 2.15. otherwise i get an error about some filter weights array? maybe not that much of a difference but just curious what this difference is

Traceback (most recent call last):
File “cryosparc_master/cryosparc_compute/jobs/rtp_workers/”, line 371, in
File “cryosparc_master/cryosparc_compute/jobs/rtp_workers/”, line 440, in
File “cryosparc_master/cryosparc_compute/jobs/rtp_workers/”, line 817, in
File “cryosparc_master/cryosparc_compute/jobs/rtp_workers/”, line 818, in
File “cryosparc_master/cryosparc_compute/jobs/ctf_estimation/”, line 70, in
File “cryosparc_master/cryosparc_compute/jobs/ctf_estimation/”, line 846, in
File “cryosparc_master/cryosparc_compute/jobs/ctf_estimation/”, line 822, in
File “/home/cryosparc_user/cryosparc2/cryosparc2_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.8/site-packages/scipy/ndimage/”, line 862, in convolve
return _correlate_or_convolve(input, weights, output, mode, cval,
File “/home/cryosparc_user/cryosparc2/cryosparc2_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.8/site-packages/scipy/ndimage/”, line 662, in _correlate_or_convolve
raise RuntimeError(‘filter weights array has incorrect shape.’)
RuntimeError: filter weights array has incorrect shape.

Hi @haraQQiri,

Interesting, thanks for reporting this. Does this problem show up in regular CryoSPARC or only in CryoSPARC live? Also, does it show up on all of your datasets or just on some?


I only see that error message in the live instance, but in normal cryosparc for the ctf job i don’t see the option for it. i also noticed that even with the classic mode in live, or just normal running in the normal the ctf fits do not seem to be working on datasets that otherwise look normal on servers running cryosparc and other software but do not return that error. hoping that whatever is triggering that message can be tracked down

What are the CryoSPARC versions on those servers?

the ones where we don’t have issues are still running 4.2.1, the one throwing this error was running an older version, though i forget which one, then updated to 4.3.0

Was there ever a solution to this problem? I am getting this error on one of my datasets on CryoSPARC while other datasets are running CTF refinement fine.

Welcome to the forum @may023.
Please can you post

  • the CryoSPARC version
  • the CryoSPARC job type
  • the full traceback of the error
  • where in CryoSPARC (event log? job log?) you observed the error