 |
 |
Hi Sjors *et al. *and all, Sorry if this is spam, it may be useful to some, the bitbucket.org/scheres/relion-3.0_beta/issues page is access denied otherwise I would have posted there. I'm struggling to successfully execute the polishing job in the relion-3.0-beta-2 release. I've recompiled after a git pull, however the problem persists. I've firstly ran *"RelionCorr" *without issues. Inspection of the output directory indicates nothing out of the ordinary. The DW summed averages are present, together with the .log, .eps and .star files. See excerpts below... *e.g. 0001_raw.star* *data_general* *_rlnImageSizeX 7420* *_rlnImageSizeY 7676* *_rlnImageSizeZ 32* *_rlnMicrographMovieName raw/0001_raw.mrcs* *_rlnMicrographBinning 2.000000* *_rlnMicrographOriginalPixelSize 0.527000* *_rlnMicrographDoseRate 1.400000* *_rlnMicrographPreExposure 0.000000* *_rlnVoltage 300.000000* *_rlnMicrographStartFrame 1* *_rlnMotionModelVersion 1* *data_global_shift* *loop_ * *_rlnMicrographFrameNumber #1 * *_rlnMicrographShiftX #2 * *_rlnMicrographShiftY #3 * * 1 0.000000 0.000000 * * 2 -0.32314 0.572730 * * 3 0.035907 1.630819 * * 4 0.629337 2.125151 * * 5 -0.16405 1.561933 * *...* *micrographs_corrected.star* *# RELION; version 3.0-beta-2* *data_* *loop_ * *_rlnMicrographName #1 * *_rlnMicrographMetadata #2 * *_rlnAccumMotionTotal #3 * *_rlnAccumMotionEarly #4 * *_rlnAccumMotionLate #5 * *MotionCorr/job003/raw/0001_raw.mrc MotionCorr/job003/raw/0001_raw.star 18.703075 0.657601 18.045474* *MotionCorr/job003/raw/0002_raw.mrc MotionCorr/job003/raw/0002_raw.star 47.114448 1.409654 45.704794* *MotionCorr/job003/raw/0003_raw.mrc MotionCorr/job003/raw/0003_raw.star 36.212216 10.730992 25.481224* *MotionCorr/job003/raw/0004_raw.mrc MotionCorr/job003/raw/0004_raw.star 37.479071 9.165227 28.313844* *...* Upon trying the bayesian polishing I've encountered the following error Command: *`which relion_motion_refine` --i Refine3D/job217/run_ct10_data.star --f PostProcess/job004/postprocess.star --corr_mic MotionCorr/job003/corrected_micrographs.star --m1 Refine3D/job217/run_ct10_half1_class001_unfil.mrc --m2 Refine3D/job217/run_ct10_half2_class001_unfil.mrc --mask MaskCreate/job172/mask.mrc --first_frame 1 --last_frame -1 --o Polish/job031/ --min_p 10000 --eval_frac 0.5 --align_frac 0.5 --params3 --only_do_unfinished --j 20 * Output: *in: /usr/local/programs/relion-3.0_beta/src/jaz/micrograph_handler.cpp, line 509* *=== Backtrace ===* */usr/local/programs/relion-3.0_beta/build/bin/relion_motion_refine(_ZN11RelionErrorC2ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES7_l+0x6d) [0x433aad]* */usr/local/programs/relion-3.0_beta/build/bin/relion_motion_refine(_ZN17MicrographHandler11getMetaNameENSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE+0x1f2) [0x52e082]* */usr/local/programs/relion-3.0_beta/build/bin/relion_motion_refine(_ZN17MicrographHandler11loadInitialERK13MetaDataTabledbRiRdRNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE+0x17d) [0x52f1ad]* */usr/local/programs/relion-3.0_beta/build/bin/relion_motion_refine(_ZN17MicrographHandler4initERK13MetaDataTabledbiRiRdRNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE+0x99) [0x530489]* */usr/local/programs/relion-3.0_beta/build/bin/relion_motion_refine(_ZN13MotionRefiner4initEv+0x4db) [0x4768ab]* */usr/local/programs/relion-3.0_beta/build/bin/relion_motion_refine(main+0x44) [0x4231e4]* */lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0) [0x7efcf7505830]* */usr/local/programs/relion-3.0_beta/build/bin/relion_motion_refine(_start+0x29) [0x423d69]* *==================* *ERROR: * *ERROR: MicrographHandler::getMetaName: no metadata star-file for raw/0005_movie.mrcs found in MotionCorr/job003/corrected_micrographs.star.* Clearly the unaligned raw movies are required, which in my case are stored in $proj_dir/raw/. However the path for these is not included in the corrected_micrograph.star file as the Error indicates. The only reference to them that I can find is in the individual 0001_raw.star etc. files in the output directory. My understanding of the new Relion implementation of MotionCor, is that the aligned movies are not written and so I assume the Bayesian polishing must search the original directory of raw movies - so this raises the question, is it a requirement that these be named XXXX_movie.mrcs? I'm not sure whether a symbolic link should have been created (and/or a corresponding _movie.star) pointing to these raw movies with the assumed suffix of _movie.mrcs. This is what MicrographHandler is apparently looking for... I have tried creating symbolic links and tried including the path in the correct_micrograph.star. Any help would be appreciated, many thanks, Charles _________________ *Charles Bayly-Jones* BSc(ScSchProg)(Hons) *PhD Candidate* *Teaching Associate* *Monash Biomedicine Discovery Institute* *Department of Biochemistry and Molecular Biology* Level 2, Building 77 23 Innovation Walk Clayton VIC 3800 Australia ######################################################################## To unsubscribe from the CCPEM list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1