We too have encountered the same problem. We had ~1200 movies with 17
frames, about 5 movies with very few frames that we ended up discarding,
and two that had 18 instead of 17 which crashed relion fairly well. You
can define a specific last frame number rather than entering "0" to use
all framesı.
Cheers,
Neil.
Neil Ranson
Director of Student Education
School of Molecular and Cellular Biology
University of Leeds
LS2 9JT. UK.
Astbury 8.108
E: [log in to unmask]
T: +44 113 343 7065
On 03/04/2014 12:20, "Sjors Scheres" <[log in to unmask]> wrote:
>We've encountered similar problems: probably related to hardware trouble
>when writing the movies originally to disc. If it's just one or two
>missing frames in a few movies you may also "rescue" those by just
>adding the last frame of it multiple times. You could use
>relion_stack_create (or any other MRC stack handling program) to do this.
>HTH, S
>
>On 04/03/2014 12:08 PM, Radostin Danev wrote:
>> Hi Sjors,
>>
>> Thank you for your quick response!
>>
>> I checked and indeed, a few movies have less frames than the rest. I
>>will
>> remove them from the .star files and try again.
>>
>> Rado
>>
>>
>> -----Original Message-----
>> From: Collaborative Computational Project in Electron cryo-Microscopy
>> [mailto:[log in to unmask]] On Behalf Of Sjors Scheres
>> Sent: 03 April 2014 12:42
>> To: [log in to unmask]
>> Subject: Re: [ccpem] Refine with movies error
>>
>> Hi Rado,
>>
>> All particles in your movie.star file have to have the same number of
>>movie
>> frames. This is currently not the case in your data. If you have
>>different
>> exposure times (and thus frame numbers) in one data set, I would
>>recommend
>> using the highest number of frames they have in common.
>>
>> If you think your data all should have the same number of frames, then
>>have
>> a close look at all your movies to double check for errors.
>>
>> HTH,
>> S
>>
>> On 04/03/2014 10:51 AM, Radostin Danev wrote:
>>> Hi everyone,
>>>
>>> I searched the archives but did not find a similar problem reported
>>> before, so here it goes ...
>>>
>>> I am trying to refine with movies. Continuing from a previous
>>> non-movie 3D refinement which used grouped particles gave me the
>>>following
>> error:
>>> .............
>>>
>>> + Slave 125 runs on host = hpcl3009.biochem.mpg.de
>>> + Slave 126 runs on host = hpcl3009.biochem.mpg.de
>>> + Slave 127 runs on host = hpcl3009.biochem.mpg.de
>>> =================
>>> Expanding current model for movie frames...
>>> MlModel::expandToMovieFrames ERROR: unequal number of frames in
>>> groupgroup_20
>>> File: src/ml_model.cpp line: 553
>>> MlModel::expandToMovieFrames ERROR: unequal number of frames in
>>> groupgroup_20
>>> File: src/ml_model.cpp line: 553
>>> MlModel::expandToMovieFrames ERROR: unequal number of frames in
>>> groupgroup_20
>>> File: src/ml_model.cpp line: 553
>>> MlModel::expandToMovieFrames ERROR: unequal number of frames in
>>> groupgroup_20
>>> File: src/ml_model.cpp line: 553
>>> ----------------------------------------------------------------------
>>> ---- MPI_ABORT was invoked on rank 51 in communicator MPI_COMM_WORLD
>>> with errorcode 1.
>>>
>>> .............
>>>
>>> Thinking that it is a problem with the grouping I then tried to
>>> continue from a 3D refinement which used the non-grouped original
>>> particles.star file.
>>> This seemed to work but then it gave the following error:
>>>
>>> ............
>>>
>>> =============================
>>> Oversampling= 1 NrHiddenVariableSamplingPoints= 3924566016
>>> OrientationalSampling= 0.234375 NrOrientations= 107600
>>> TranslationalSampling= 1.125 NrTranslations= 196
>>> =============================
>>> Expectation iteration 34
>>> 1.02/13.72 hrs ....~~(,_,"> exp_thisparticle_sumweight= -nan [oo]
>>> exp_part_id= 13697exp_iimage=11
>>> group_id= 87 mymodel.scale_correction[group_id]= 0.831255
>>> exp_ipass= 0
>>> sampling.NrDirections(0, true)= 3259 sampling.NrDirections(0,
>>>false)=
>> 535
>>> sampling.NrPsiSamplings(0, true)= 768 sampling.NrPsiSamplings(0,
>>>false)=
>> 26
>>> mymodel.sigma2_noise[exp_ipart]=
>>> 0.04963
>>> 0.001015
>>> (skipped a lot of numbers)
>>> 9.566e-06
>>> 8.995e-06
>>>
>>> wsum_model.sigma2_noise[exp_ipart]=
>>> 0
>>> 0
>>> (skipped a lot of numbers) 0
>>> 0
>>> 0
>>>
>>> written out Mweight.spi
>>> exp_thisparticle_sumweight= -nan
>>> exp_min_diff2[exp_ipart]= 9.9e+100
>>> slave 125 encountered error: ERROR!!! zero sum of weights....
>>> File: src/ml_optimiser.cpp line: 3929
>>> +++ RELION: command line arguments (with defaults for optional ones
>>> +++ between
>>> parantheses) +++
>>>
>>> ..........
>>>
>>> What could be the problem?
>>>
>>> Thanks!
>>>
>>> Rado
>> --
>> Sjors Scheres
>> MRC Laboratory of Molecular Biology
>> Francis Crick Avenue, Cambridge Biomedical Campus Cambridge CB2 0QH,
>>U.K.
>> tel: +44 (0)1223 267061
>> http://www2.mrc-lmb.cam.ac.uk/groups/scheres
>
>--
>Sjors Scheres
>MRC Laboratory of Molecular Biology
>Francis Crick Avenue, Cambridge Biomedical Campus
>Cambridge CB2 0QH, U.K.
>tel: +44 (0)1223 267061
>http://www2.mrc-lmb.cam.ac.uk/groups/scheres
|