Dear Fuxing,
You either have 1) too low SNR in your data and you cannot refine your
model, or 2) your initial model lies outside the radius of convergence.
Often the quality of the 2D classes can help you decide which is the
case. If the 2D class averages look like blobs, then 1) is likely the
case. If the 2D classes show protein-like internal features, then 2) is
likely the case.
HTH,
Sjors
On 10/02/2016 11:14 PM, SUBSCRIBE CCPEM Fuxing Zeng wrote:
> Hi all,
>
> I'm running the 3D auto-refine for my data and it shows the WARNING about the angular accuracy is worse than 10 degrees. I'm just wondering:
> 1. What cause the accuracy worse than 10˚?
> 2. I checked the map for iteration6, even the shape is not correct. Should I stop the job?
> 3. To tune resolution by adjusting T, should I run 3D-classification after refinement or skip refinement just run 3D-classification with a single class.
>
> Thanks,
> Fuxing
>
> Auto-refine: Resolution= 38.176 (no gain for 2 iter)
> Auto-refine: Changes in angles= 29.0102 degrees; and in offsets= 1.14429 pixels (no gain for 0 iter)
> Estimating accuracies in the orientational assignment ...
> 4/ 4 sec ............................................................~~(,_,">
> Auto-refine: Estimated accuracy angles= 30.1 degrees; offsets= 10.1 pixels
> Auto-refine: WARNING: The angular accuracy is worse than 10 degrees, so basically you cannot align your particles (yet)!
> Auto-refine: WARNING: You probably need not worry if the accuracy improves during the next few iterations.
> Auto-refine: WARNING: However, if the problem persists it may lead to spurious FSC curves, so be wary of inflated resolution estimates...
> Auto-refine: WARNING: Sometimes it is better to tune resolution yourself by adjusting T in a 3D-classification with a single class.
> Auto-refine: Angular step= 7.5 degrees; local searches= false
> Auto-refine: Offset search range= 5 pixels; offset step= 1 pixels
> CurrentResolution= 38.176 Angstroms, which requires orientationSampling of at least 12.8571 degrees for a particle of diameter 340 Angstroms
> Oversampling= 0 NrHiddenVariableSamplingPoints= 96768
> OrientationalSampling= 15 NrOrientations= 4608
> TranslationalSampling= 2 NrTranslations= 21
> =============================
> Oversampling= 1 NrHiddenVariableSamplingPoints= 3096576
> OrientationalSampling= 7.5 NrOrientations= 36864
> TranslationalSampling= 1 NrTranslations= 84
> =============================
> Estimated memory for expectation step > 0.318263 Gb, available memory = 40 Gb.
> Estimated memory for maximization step > 0.0255431 Gb, available memory = 40 Gb.
--
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
|