search

I2PC / scipion-em-xmipp

Plugin to use Xmipp programs within the Scipion framework
GNU General Public License v3.0
11 stars 4 forks source link

Protocol 'xmipp - subtract projection' #356

Closed JuhaHuiskonen closed 3 years ago

JuhaHuiskonen commented 3 years ago

I am trying to run 'xmipp - subtract projection' in Scipion 3.0.7. It fails with the following error:

00001: RUNNING PROTOCOL ----------------- 00002: Hostname: xxxxxx 00003: PID: 8268 00004: pyworkflow: 3.0.8 00005: plugin: orphan 00006: currentDir: xxxxxx 00007: workingDir: Runs/005763_XmippProtSubtractProjection 00008: runMode: Restart 00009: MPI: 1 00010: threads: 64 00011: 'NoneType' object has no attribute 'Plugin' LegacyProtocol installation couldn't be validated. Possible cause could be a configuration issue. Try to run scipion config. 00012: Protocol has validation errors: 00013: 'NoneType' object has no attribute 'Plugin' LegacyProtocol installation couldn't be validated. Possible cause could be a configuration issue. Try to run scipion config. 00014: ------------------- PROTOCOL FAILED (DONE 0/0)

rmarabini commented 3 years ago

Hi @DStrelak, please delete "'xmipp - subtract projection" it does not make sense to maintain it. I guess @JuhaHuiskonen is trying to use it because "relion subtract projection" does not longer work with an arbitrary set of particles. I am modifying "relion subtract projection" so it has all the functionally available in Scipion 2 version. I will let you know when I am done.

JuhaHuiskonen commented 3 years ago

Hi @rmarabini indeed I would use 'relion - subtract projection', if it worked with subsets of particles.

DStrelak commented 3 years ago

Hi @JuhaHuiskonen , is this a valid solution for you? @rmarabini , do we have any workaround meanwhile?

JuhaHuiskonen commented 3 years ago

Hi @DStrelak yes, if the relion version can be made to work with any set of particles then it would be a valid solution for me (my dataset is so large that a single job requires more than the time limit on the cluster - so the only option seems to be to split the SetOfParticles from RELION).

DStrelak commented 3 years ago

my dataset is so large that a single job requires more than the time limit on the cluster

Have you tried / is it applicable in your case to use pwem-subset / pwem-split sets protocol? There is also xmipp_extract_subset program, which, unlike the split protocol, also extracts the particles from the stack file (i.e. you don't need the original stack file with all particles with your newly created smaller metadata file).

rmarabini commented 3 years ago

I do not think we have any alternative.

On Thu, Feb 4, 2021 at 11:49 AM David Strelak notifications@github.com wrote:

Hi @JuhaHuiskonen https://github.com/JuhaHuiskonen , is this a valid solution for you? @rmarabini https://github.com/rmarabini , do we have any workaround meanwhile?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/I2PC/scipion-em-xmipp/issues/356#issuecomment-773215066, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACQGTIXNZL6HYBJST5GIQ5LS5J3TBANCNFSM4XBV62UA .

JuhaHuiskonen commented 3 years ago

@DStrelak I have tried splitting. But subsets are not compatible with relion subtraction, as its output is relion - auto refine 3D protocol, not a SetOfParticles produced by subset/split.