Re: [scipion-users] xmipp3- ctf--> NonAstigmaticValidity-->2d classification ctf correction

[Carlos O. S. <co...@cn...>]

Dear Dmitry,

in principle correcting for the astigmatism while reconstructing could 
be possible. In practice, the angle of the defoci in the micrograph 
should be taken into account for this correction and no 3D 
reconstruction program is currently considering it, so it is better to 
keep low the astigmatism.

The main parameters to determine the quality of the micrographs are:

- ctfCritMaxFreq: in Angstroms which is the maximum resolution expected 
due to envelope

- ctfCritNonAstigmaticValidity: in Angstroms which is the maximum 
resolution due to astigmatism

- ctfCritCtfMargin and ctfCritCorr13: the higher, the better. It is 
related to how visible Thon rings are

- ctfCritPSDCorr90: it can detect drift and astigmatism, should be very 
close to 1 for good micrographs

- ctfCritFirstZeroRatio: it can detect astigmatism. It should be very 
close to 1 for good micrographs

Kind regards, Carlos Oscar


On 15/09/2016 15:53, Dmitry A. Semchonok wrote:
> Dear colleagues,
>
> My question is about xmipp3- ctf estimation.
>
> In that protocol we have the NonAstigmaticValidity parameter.
> I would like to ask if knowing this parameter later the protocol relion 2d
> (3d)classification is correcting for that astigmatism or it is better to
> remove the images where this NonAstigmaticValidity is quite high (higher
> the resolution you expect)?
>
> Which parameter in xmipp3-ctf by your experience is the most crucial and
> requires to kick the bad images out?
>
> Sincerely,
> Dmitry
>
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> scipion-users mailing list
> sci...@li...
> https://lists.sourceforge.net/lists/listinfo/scipion-users
>

-- 
------------------------------------------------------------------------
Carlos Oscar Sánchez Sorzano                  e-mail:   co...@cn...
Biocomputing unit                             http://biocomp.cnb.csic.es
National Center of Biotechnology (CSIC)
c/Darwin, 3
Campus Universidad Autónoma (Cantoblanco)     Tlf: 34-91-585 4510
28049 MADRID (SPAIN)                          Fax: 34-91-585 4506
------------------------------------------------------------------------