 |
 |
Hi, Did you change the pixel size during processing? For Bayesian Polishing, three pixel sizes are important: (1) original movie pixel size (2) binned pixel size after motion correction (3) extracted particle pixel size These are stored in movie metadata STAR files and particle STAR files. If any of them are wrong or inconsistent, polishing will produce rubbish. Please try Class2D to make sure your polished particles are fine. Best regards, Takanori Nakane > Dear Takanori > > I could only run the polish with default parameters as the training mode > does not work. I encounter the following problem now: > the refinement is super slow (30-48h/iteration) and the changes in angle > stays at 20 all the time (whereas in unpolished is at 5 at the same > iteration). Do you have a suggestion? or you think it is normal? > > *POLISHED:* > Auto-refine: Iteration= 8 > Auto-refine: Resolution= 12.871 (no gain for 1 iter) > Auto-refine: Changes in angles= 23.253 degrees; and in offsets= 1.65378 > pixels (no gain for 0 iter) > Estimating accuracies in the orientational assignment ... > 3/ 3 sec > ............................................................~~(,_,"> > Auto-refine: Estimated accuracy angles= 3.748 degrees; offsets= 3.058 > pixels > Auto-refine: Angular step= 3.75 degrees; local searches= false > Auto-refine: Offset search range= 5 pixels; offset step= 1 pixels > CurrentResolution= 12.871 Angstroms, which requires orientationSampling > of at least 4.09091 degrees for a particle of diameter 360 Angstroms > Oversampling= 0 NrHiddenVariableSamplingPoints= 774144 > OrientationalSampling= 7.5 NrOrientations= 36864 > TranslationalSampling= 2 NrTranslations= 21 > ============================= > Oversampling= 1 NrHiddenVariableSamplingPoints= 24772608 > OrientationalSampling= 3.75 NrOrientations= 294912 > TranslationalSampling= 1 NrTranslations= 84 > ============================= > Expectation iteration 8 > 16.41/30.47 hrs ................................~~(,_,"> > > > *UNPOLISHED:* > Auto-refine: Iteration= 8 > Auto-refine: Resolution= 11.7353 (no gain for 0 iter) > Auto-refine: Changes in angles= 4.67816 degrees; and in offsets= > 0.998428 pixels (no gain for 0 iter) > Estimating accuracies in the orientational assignment ... > 3/ 3 sec > ............................................................~~(,_,"> > Auto-refine: Estimated accuracy angles= 2.117 degrees; offsets= 1.612 > pixels > Auto-refine: Angular step= 3.75 degrees; local searches= false > Auto-refine: Offset search range= 5 pixels; offset step= 1 pixels > CurrentResolution= 11.7353 Angstroms, which requires orientationSampling > of at least 3.71134 degrees for a particle of diameter 360 Angstroms > Oversampling= 0 NrHiddenVariableSamplingPoints= 774144 > OrientationalSampling= 7.5 NrOrientations= 36864 > TranslationalSampling= 2 NrTranslations= 21 > ============================= > Oversampling= 1 NrHiddenVariableSamplingPoints= 24772608 > OrientationalSampling= 3.75 NrOrientations= 294912 > TranslationalSampling= 1 NrTranslations= 84 > ============================= > Expectation iteration 8 > 1.72/1.72 hrs > ............................................................~~(,_,"> > > ######################################################################## To unsubscribe from the CCPEM list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1