 |
 |
Hi Domen and Zhaoming, I forward comments from Jasenko Zivanov (CC-ed), the developer of Bayesian Polishing. Please make sure all FCC files under the job directory have the same size. > Could it be that some of the (thousands of) partial FCC files (*_FCC_cc.mrc, *_FCC_w0/1.mrc) are damaged? > > Sjors changed the mrc-reader afterwards, so that it would refuse to load a corrupted image. > That problem should be gone in the current version. > > I had observed a corrupted partial FCC file once myself. > This could definitely introduce NaNs into the FCC computation. > > The way to diagnose this would be to check if there are unusually small files in the output-directory. > All of them should be the same size. Best regards, Takanori Nakane On 2018/09/28 17:34, Zhaoming Su wrote: > Hi All, > > I have the exact problem on multiple data sets collected as packed mrc from EPU, except they all fail at the second frame... > > # RELION; version 3.0-beta-2 > > data_perframe_bfactors > > loop_ > _rlnMovieFrameNumber #1 > _rlnBfactorUsedForSharpening #2 > _rlnFittedInterceptGuinierPlot #3 > 0 -0.04828 6.304424e-05 > 1 -5.15980e+10 -inf > 2 -5.15980e+10 -inf > 3 -5.15980e+10 -inf > 4 -5.15980e+10 -inf > 5 -5.15980e+10 -inf > 6 -5.15980e+10 -inf > 7 -5.15980e+10 -inf > 8 -5.15980e+10 -inf > 9 -5.15980e+10 -inf > 10 -5.15980e+10 -inf > 11 -5.15980e+10 -inf > 12 -5.15980e+10 -inf > 13 -5.15980e+10 -inf > 14 -5.15980e+10 -inf > 15 -5.15980e+10 -inf > 16 -5.15980e+10 -inf > 17 -5.15980e+10 -inf > 18 -5.15980e+10 -inf > 19 -5.15980e+10 -inf > 20 -5.15980e+10 -inf > 21 -5.15980e+10 -inf > 22 -5.15980e+10 -inf > 23 -5.15980e+10 -inf > 24 -5.15980e+10 -inf > > > > Best, > > > Zhaoming Su, Ph.D. > > Research Scientist, Bioengineering > > Stanford University > > ________________________________ > From: Collaborative Computational Project in Electron cryo-Microscopy <[log in to unmask]> on behalf of Domen Kampjut <[log in to unmask]> > Sent: Friday, September 28, 2018 7:45:40 AM > To: [log in to unmask] > Subject: [ccpem] Bayesian Polishing b-factor fitting error > > Dear All, > > I keep running into an error when trying to polish particles in > relion3.0. Polishing job starts normally and estimated per-particle > motion for all micrographs look very reasonable judging from the .eps > files in the Micrographs subfolder. Estimating B-factors fails after > 10th frame though (see the bfactors.star output below) and the job > terminates with segmantation fault. I only see this error in the dataset > collected with a K2 camera as packed .mrc movies using EPU. Polishing of > Falcon-III movies works fine. Also, if I limit the polishing job to the > first 9 frames the job completes normally. Please let me know if anyone > else observed the same error and what you would need to reproduce it. > > Many thanks and best wishes, > > Domen Kampjut > > > > -------------------------------------------------------------- > > data_perframe_bfactors > > loop_ > _rlnMovieFrameNumber #1 > _rlnBfactorUsedForSharpening #2 > _rlnFittedInterceptGuinierPlot #3 > 0 -10.90545 0.048441 > 1 -0.08144 -0.21434 > 2 -0.08144 -0.26051 > 3 -3.19811 -0.29255 > 4 -5.69430 -0.29903 > 5 -12.85827 -0.27141 > 6 -19.60021 -0.24128 > 7 -22.26604 -0.23126 > 8 -28.06880 -0.19417 > 9 -30.55454 -0.11258 > 10 -8.80873e+10 -inf > 11 -8.80873e+10 -inf > 12 -8.80873e+10 -inf > 13 -8.80873e+10 -inf > 14 -8.80873e+10 -inf > 15 -8.80873e+10 -inf > 16 -8.80873e+10 -inf > 17 -8.80873e+10 -inf > 18 -8.80873e+10 -inf > 19 -8.80873e+10 -inf > 20 -8.80873e+10 -inf > 21 -8.80873e+10 -inf > 22 -8.80873e+10 -inf > 23 -8.80873e+10 -inf > 24 -8.80873e+10 -inf > 25 -8.80873e+10 -inf > 26 -8.80873e+10 -inf > 27 -8.80873e+10 -inf > 28 -8.80873e+10 -inf > 29 -8.80873e+10 -inf > 30 -8.80873e+10 -inf > 31 -8.80873e+10 -inf > 32 -8.80873e+10 -inf > 33 -8.80873e+10 -inf > 34 -8.80873e+10 -inf > 35 -8.80873e+10 -inf > 36 -8.80873e+10 -inf > 37 -8.80873e+10 -inf > 38 -8.80873e+10 -inf > 39 -8.80873e+10 -inf > > ######################################################################## > > To unsubscribe from the CCPEM list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1 > > ######################################################################## > > To unsubscribe from the CCPEM list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1 > ######################################################################## To unsubscribe from the CCPEM list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1