Bayesian polishing with EER file format.

[DmitrySemchonok]

Describe the bug When running the relion- bayesian polishing, the process crashes. Original files are in EER format.

To Reproduce Steps to reproduce the behaviour:

0. run the bayesian polishing training with default 10 000 particles and use the parameters for the bayesian polishing In my case Output params: Sigma for velocity: 1.140 Sigma for divergence: 2745.0 Sigma for acceleration: 3.00

1. Bayesian polishing protocol: Set as the input motion-movie alignment files. I used motioncor plugin v: 3.7

2. Use the setting from step 0.

3. Keep the other parameters as default

4. Set any parameters of Thread/MPI that your system allows. (I tested most of the combinations including Threads = 1, MPI = 1)

5. Click on 'Execute'

6. The process starts normally.

7. After some time - get an error.

Expected behavior Bayesian polishing protocol to finish without errors.

Screenshots If applicable

Logs Attach run.stderr log file from the failing protocol. Traceback (most recent call last): File "/home/user/miniconda3/envs/scipion3/lib/python3.8/site-packages/pyworkflow/protocol/protocol.py", line 202, in run self._run() File "/home/user/miniconda3/envs/scipion3/lib/python3.8/site-packages/pyworkflow/protocol/protocol.py", line 253, in _run resultFiles = self._runFunc() File "/home/user/miniconda3/envs/scipion3/lib/python3.8/site-packages/pyworkflow/protocol/protocol.py", line 249, in _runFunc return self._func(*self._args) File "/home/user/miniconda3/envs/scipion3/lib/python3.8/site-packages/relion/protocols/protocol_bayesian_polishing.py", line 371, in trainOrPolishStep self.runJob(self._getProgram('relion_motion_refine'), args) File "/home/user/miniconda3/envs/scipion3/lib/python3.8/site-packages/pyworkflow/protocol/protocol.py", line 1475, in runJob self._stepsExecutor.runJob(self._log, program, arguments, **kwargs) File "/home/user/miniconda3/envs/scipion3/lib/python3.8/site-packages/pyworkflow/protocol/executor.py", line 65, in runJob process.runJob(log, programName, params, File "/home/user/miniconda3/envs/scipion3/lib/python3.8/site-packages/pyworkflow/utils/process.py", line 55, in runJob return runCommand(command, env, cwd) File "/home/user/miniconda3/envs/scipion3/lib/python3.8/site-packages/pyworkflow/utils/process.py", line 70, in runCommand check_call(command, shell=True, stdout=sys.stdout, stderr=sys.stderr, File "/home/user/miniconda3/envs/scipion3/lib/python3.8/subprocess.py", line 364, in check_call raise CalledProcessError(retcode, cmd) subprocess.CalledProcessError: Command ' mpirun -np 5 which relion_motion_refine_mpi --i Runs/003535_ProtRelionBayesianPolishing/input_particles.star --o Runs/003535_ProtRelionBayesianPolishing/extra --f Runs/003232_ProtRelionPostprocess/extra/postprocess.star --angpix_ref 0.45450 --corr_mic Runs/003535_ProtRelionBayesianPolishing/input_corrected_micrographs.star --first_frame 1 --last_frame 0 --s_vel 1.140 --s_div 2745.000 --s_acc 3.000 --bfac_minfreq 20.000 --bfac_maxfreq -1.000 --combine_frames --j 5 ' returned non-zero exit status 139. [31mProtocol failed: Command ' mpirun -np 5 which relion_motion_refine_mpi --i Runs/003535_ProtRelionBayesianPolishing/input_particles.star --o Runs/003535_ProtRelionBayesianPolishing/extra --f Runs/003232_ProtRelionPostprocess/extra/postprocess.star --angpix_ref 0.45450 --corr_mic Runs/003535_ProtRelionBayesianPolishing/input_corrected_micrographs.star --first_frame 1 --last_frame 0 --s_vel 1.140 --s_div 2745.000 --s_acc 3.000 --bfac_minfreq 20.000 --bfac_maxfreq -1.000 --combine_frames --j 5 ' returned non-zero exit status 139. [0m

Desktop (please complete the following information):

• Scipion version: v3.0.11-Eugenios - 3.0.25
• Relion plugin version 4.0.8

Additional context Add any other context about the problem here.

[azazellochg]

Thank you for the report @DmitrySemchonok . I will need more information: 1) first 100 lines of Runs/003535_ProtRelionBayesianPolishing/input_corrected_micrographs.star 2) you didnt attach the complete run.stderr

[DmitrySemchonok]

Thank you for the report @DmitrySemchonok . I will need more information:

1. first 100 lines of Runs/003535_ProtRelionBayesianPolishing/input_corrected_micrographs.star

2. you didnt attach the **complete** run.stderr

Because the website does not accept loading the file type. I will add it additionally.

[azazellochg]

The file can be renamed to txt

[DmitrySemchonok]

The file can be renamed to txt

1. input_corrected_micrographs.txt

2. run.txt

[azazellochg]

thanks! 1) it looks like you have not run assign optics protocol, hence you are missing the gain reference which is required for EER data. 2) my mistake, I keep forgetting that stderr will not have stdout lines...

[azazellochg]

Could you post 100 lines of Runs/003535_ProtRelionBayesianPolishing/input/GridSquare_1783832_Data_FoilHole_5969362_Data_5966229_5966231_20220906_180402_EER.star ? Maybe the gain is there?

[azazellochg]

Also, could you attach stdout as well please?

[DmitrySemchonok]

Could you post 100 lines of Runs/003535_ProtRelionBayesianPolishing/input/GridSquare_1783832_Data_FoilHole_5969362_Data_5966229_5966231_20220906_180402_EER.star ? Maybe the gain is there? GridSquare_1783832_Data_FoilHole_5969362_Data_5966229_5966231_20220906_180402_EER.txt

[DmitrySemchonok]

Also, could you attach stdout as well please?

run_out.txt

[azazellochg]

@DmitrySemchonok 1) you need to run assign optics protocol on the movies produced by mc2 protocol and set the gain reference correctly. 2) considering that it worked for 3h and not crashed immediately, it might be smth else.

[azazellochg]

Worked after the gain was assigned.