Menu
▾
▴
Re: [scipion-users] relion - 2D classification failed (done 2/3)
|
From: Carlos O. S. <co...@cn...> - 2018-05-05 07:17:56
|
Dear all, this problem normally appears after rescaling the images. You may solve it by "preprocess particles" and renormalizing the images again after rescaling. Kind regards, Carlos Oscar On 03/05/2018 16:28, Joseph Ho wrote: > Hi, Jose: > > I ran relion particle extraction.. still not working. I also don't do > downscaling in the Xmipp but still not working. > Only I use --dont_check_norm. Then the relion 2D classification process. > Not sure why all the normalise doesn't work in my Ximpp/relion > particle extraction? > > > Thanks for your help. > > Joseph > > On Thu, May 3, 2018 at 7:22 PM, Jose Miguel de la Rosa Trevin > <del...@gm...> wrote: >> Hi Joseph, >> >> Have you extracted the particles with xmipp protocol? >> Can you try extracting with relion and check if the error remains? >> I think the relion - extract particles protocol is not in the menu (we need >> to fix that >> for the upcoming bugfix release) but you can find it with Ctrl + F in the >> searching dialog. >> >> If the relion 2D jobs still complains about the normalization when using the >> particles >> extracted with relion (I have observed this when downscaling). You can try >> to use >> the advanced option '--dont_check_norm' >> >> Hope this helps, >> Jose Miguel >> >> >> On Thu, May 3, 2018 at 12:49 PM, Joseph Ho <sbd...@gm...> wrote: >>> Hi, Grigory: >>> >>> Thanks for your reply and help. I notice that normalised error. I also >>> puzzle about this. Because I do click the normalise function during >>> the particle extraction of scipion. I followed the scipion youtube >>> (beta-gal) setup. Normalise type:Ramp; Background radius -1, >>> >>> Thanks for your help >>> >>> Joseph >>> >>> On Thu, May 3, 2018 at 6:19 PM, Gregory Sharov <sha...@gm...> >>> wrote: >>>> Hello Joseph, >>>> >>>> the error message states precisely what is wrong with your job: >>>> >>>> 00341: ERROR: It appears that these images have not been normalised to >>>> an >>>> average background value of 0 and a stddev value of 1. >>>> 00342: Note that the average and stddev >>>> values for the background are calculated: >>>> 00343: (1) for single particles: outside >>>> a >>>> circle with the particle diameter >>>> 00344: (2) for helical segments: outside >>>> a >>>> cylinder (tube) with the helical tube diameter >>>> 00345: You can use the relion_preprocess >>>> program to normalise your images >>>> 00346: If you are sure you have >>>> normalised >>>> the images correctly (also see the RELION Wiki), you can switch off this >>>> error message using the --dont_check_norm command line option >>>> >>>> >>>> Best regards, >>>> Grigory >>>> >>>> >>>> -------------------------------------------------------------------------------- >>>> Grigory Sharov, Ph.D. >>>> >>>> MRC Laboratory of Molecular Biology, >>>> Francis Crick Avenue, >>>> Cambridge Biomedical Campus, >>>> Cambridge CB2 0QH, UK. >>>> tel. +44 (0) 1223 267542 >>>> e-mail: gs...@mr... >>>> >>>> On Thu, May 3, 2018 at 4:02 AM, Joseph Ho <jo...@ga...> >>>> wrote: >>>>> Dear Sir: >>>>> >>>>> >>>>> >>>>> Relion 2D was running but I stopped it. After I tried to restart, I got >>>>> this error msg. I could not get Relion 2D running anymore. I also tried >>>>> MPI >>>>> is 1 but it still failed. I use scipion v.1.2 and I installed it from >>>>> source. >>>>> >>>>> >>>>> >>>>> The msg from run.stdout >>>>> >>>>> >>>>> >>>>> 00309: The deprecated forms *will* disappear in a future version of >>>>> Open >>>>> MPI. >>>>> 00310: Please update to the new syntax. >>>>> 00311: >>>>> >>>>> -------------------------------------------------------------------------- >>>>> 00312: >>>>> >>>>> -------------------------------------------------------------------------- >>>>> 00313: [[50020,1],0]: A high-performance Open MPI point-to-point >>>>> messaging module >>>>> 00314: was unable to find any relevant network interfaces: >>>>> 00315: >>>>> 00316: Module: OpenFabrics (openib) >>>>> 00317: Host: spgpu3 >>>>> 00318: >>>>> 00319: Another transport will be used instead, although this may >>>>> result >>>>> in >>>>> 00320: lower performance. >>>>> 00321: >>>>> >>>>> -------------------------------------------------------------------------- >>>>> 00322: === RELION MPI setup === >>>>> 00323: + Number of MPI processes = 3 >>>>> 00324: + Master (0) runs on host = spgpu3 >>>>> 00325: + Slave 1 runs on host = spgpu3 >>>>> 00326: + Slave 2 runs on host = spgpu3 >>>>> 00327: ================= >>>>> 00328: uniqueHost spgpu3 has 2 ranks. >>>>> 00329: Slave 1 will distribute threads over devices 1 2 3 >>>>> 00330: Thread 0 on slave 1 mapped to device 1 >>>>> 00331: Slave 2 will distribute threads over devices 1 2 3 >>>>> 00332: Thread 0 on slave 2 mapped to device 1 >>>>> 00333: Device 1 on spgpu3 is split between 2 slaves >>>>> 00334: Running CPU instructions in double precision. >>>>> 00335: [spgpu3:17287] 2 more processes have sent help message >>>>> help-mpi-btl-base.txt / btl:no-nics >>>>> 00336: [spgpu3:17287] Set MCA parameter "orte_base_help_aggregate" to >>>>> 0 >>>>> to see all help / error messages >>>>> 00337: + WARNING: Changing psi sampling rate (before oversampling) >>>>> to >>>>> 5.625 degrees, for more efficient GPU calculations >>>>> 00338: Estimating initial noise spectra >>>>> 00339: 1/ 20 sec ..~~(,_,"> fn_img= >>>>> 067786@Runs/000695_XmippProtCropResizeParticles/extra/output_images.stk >>>>> bg_avg= 0.429345 bg_stddev= 0.852297 bg_radius= 14.0174 >>>>> 00340: ERROR: >>>>> 00341: ERROR: It appears that these images have not been normalised >>>>> to >>>>> an average background value of 0 and a stddev value of 1. >>>>> 00342: Note that the average and stddev >>>>> values for the background are calculated: >>>>> 00343: (1) for single particles: outside >>>>> a >>>>> circle with the particle diameter >>>>> 00344: (2) for helical segments: outside >>>>> a >>>>> cylinder (tube) with the helical tube diameter >>>>> 00345: You can use the relion_preprocess >>>>> program to normalise your images >>>>> 00346: If you are sure you have >>>>> normalised >>>>> the images correctly (also see the RELION Wiki), you can switch off >>>>> this >>>>> error message using the --dont_check_norm command line option >>>>> 00347: File: >>>>> /usr/local/scipion/software/em/relion-2.1/src/ml_optimiser.cpp line: >>>>> 1879 >>>>> 00348: >>>>> >>>>> -------------------------------------------------------------------------- >>>>> 00349: MPI_ABORT was invoked on rank 2 in communicator MPI_COMM_WORLD >>>>> 00350: with errorcode 1. >>>>> 00351: >>>>> 00352: NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI >>>>> processes. >>>>> 00353: You may or may not see output from other processes, depending >>>>> on >>>>> 00354: exactly when Open MPI kills them. >>>>> 00355: >>>>> >>>>> -------------------------------------------------------------------------- >>>>> 00356: Traceback (most recent call last): >>>>> 00357: File "/usr/local/scipion/pyworkflow/protocol/protocol.py", >>>>> line >>>>> 186, in run >>>>> 00358: self._run() >>>>> 00359: File "/usr/local/scipion/pyworkflow/protocol/protocol.py", >>>>> line >>>>> 233, in _run >>>>> 00360: resultFiles = self._runFunc() >>>>> 00361: File "/usr/local/scipion/pyworkflow/protocol/protocol.py", >>>>> line >>>>> 229, in _runFunc >>>>> 00362: return self._func(*self._args) >>>>> 00363: File >>>>> "/usr/local/scipion/pyworkflow/em/packages/relion/protocol_base.py", >>>>> line >>>>> 880, in runRelionStep >>>>> 00364: self.runJob(self._getProgram(), params) >>>>> 00365: File "/usr/local/scipion/pyworkflow/protocol/protocol.py", >>>>> line >>>>> 1138, in runJob >>>>> 00366: self._stepsExecutor.runJob(self._log, program, arguments, >>>>> **kwargs) >>>>> 00367: File "/usr/local/scipion/pyworkflow/protocol/executor.py", >>>>> line >>>>> 56, in runJob >>>>> 00368: env=env, cwd=cwd) >>>>> 00369: File "/usr/local/scipion/pyworkflow/utils/process.py", line >>>>> 51, >>>>> in runJob >>>>> 00370: return runCommand(command, env, cwd) >>>>> 00371: File "/usr/local/scipion/pyworkflow/utils/process.py", line >>>>> 65, >>>>> in runCommand >>>>> 00372: check_call(command, shell=True, stdout=sys.stdout, >>>>> stderr=sys.stderr, env=env, cwd=cwd) >>>>> 00373: File >>>>> "/usr/local/scipion/software/lib/python2.7/subprocess.py", >>>>> line 186, in check_call >>>>> 00374: raise CalledProcessError(retcode, cmd) >>>>> 00375: CalledProcessError: Command 'mpirun -np 3 -bynode `which >>>>> relion_refine_mpi` --gpu 1,2,3 --tau2_fudge 2 --scale >>>>> --dont_combine_weights_via_disc --iter 25 --norm --psi_step 10.0 >>>>> --ctf >>>>> --offset_range 5.0 --oversampling 1 --pool 3 --o >>>>> Runs/000827_ProtRelionClassify2D/extra/relion --i >>>>> Runs/000827_ProtRelionClassify2D/input_particles.star >>>>> --particle_diameter >>>>> 200 --K 200 --flatten_solvent --zero_mask --offset_step 2.0 --angpix >>>>> 7.134 >>>>> --j 1' returned non-zero exit status 1 >>>>> 00376: Protocol failed: Command 'mpirun -np 3 -bynode `which >>>>> relion_refine_mpi` --gpu 1,2,3 --tau2_fudge 2 --scale >>>>> --dont_combine_weights_via_disc --iter 25 --norm --psi_step 10.0 >>>>> --ctf >>>>> --offset_range 5.0 --oversampling 1 --pool 3 --o >>>>> Runs/000827_ProtRelionClassify2D/extra/relion --i >>>>> Runs/000827_ProtRelionClassify2D/input_particles.star >>>>> --particle_diameter >>>>> 200 --K 200 --flatten_solvent --zero_mask --offset_step 2.0 --angpix >>>>> 7.134 >>>>> --j 1' returned non-zero exit status 1 >>>>> 00377: FAILED: runRelionStep, step 2 >>>>> 00378: 2018-05-03 10:35:36.970022 >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> Thanks for your help >>>>> >>>>> >>>>> >>>>> Meng-Chiao (Joseph) Ho >>>>> >>>>> Assistant Research Fellow >>>>> >>>>> Institute of Biological Chemistry, Academia Sinica >>>>> >>>>> No. 128, Sec 2, Academia Road, Nankang, Taipei 115, Taiwan >>>>> >>>>> Tel: 886-2-27855696 ext 3080/3162 >>>>> >>>>> Email: jo...@ga... >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------------------ >>>>> Check out the vibrant tech community on one of the world's most >>>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot >>>>> _______________________________________________ >>>>> scipion-users mailing list >>>>> sci...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/scipion-users >>>>> >>>> >>>> >>>> ------------------------------------------------------------------------------ >>>> Check out the vibrant tech community on one of the world's most >>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot >>>> _______________________________________________ >>>> scipion-users mailing list >>>> sci...@li... >>>> https://lists.sourceforge.net/lists/listinfo/scipion-users >>>> >>> >>> ------------------------------------------------------------------------------ >>> Check out the vibrant tech community on one of the world's most >>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot >>> _______________________________________________ >>> scipion-users mailing list >>> sci...@li... >>> https://lists.sourceforge.net/lists/listinfo/scipion-users >> >> >> ------------------------------------------------------------------------------ >> Check out the vibrant tech community on one of the world's most >> engaging tech sites, Slashdot.org! http://sdm.link/slashdot >> _______________________________________________ >> scipion-users mailing list >> sci...@li... >> https://lists.sourceforge.net/lists/listinfo/scipion-users >> > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > scipion-users mailing list > sci...@li... > https://lists.sourceforge.net/lists/listinfo/scipion-users > -- ------------------------------------------------------------------------ Carlos Oscar Sánchez Sorzano e-mail: co...@cn... Biocomputing unit http://i2pc.es/coss National Center of Biotechnology (CSIC) c/Darwin, 3 Campus Universidad Autónoma (Cantoblanco) Tlf: 34-91-585 4510 28049 MADRID (SPAIN) Fax: 34-91-585 4506 ------------------------------------------------------------------------ |
