[3dem] [ccpem] on FSC curve (A can of worms...)

[Jose Maria Carazo]

My god, I do not know how we get into these debates by email.....

Just on Ed email, indeed it is a measure of reproducibility, calculated by
splitting the data sets into two parts...... However, have you thought how
many different ways one has of splitting a data set of N samples into two
of N/2 samples?. Quite a few. Naturally, this introduces another layer of
statistics in the interpretation of that measure. We did some simple tests
with very reduced data sets in the work below. Surely it could be done much
better now

With best wishes..JM

L. G. de la Fraga, J. Dopazo and J. M. Carazo. Confidence Limits for
Resolution Estimation in Image Averaging by Random Subsamplig,
Ultramicroscopy 60 (1995) pp. 385-391

On Wed, Aug 12, 2015 at 3:26 PM, Edward Egelman <egelman at virginia.edu>
wrote:

> Since Marin brought up the can of worms metaphor, let me just add my own
> two bits: FSC has never been a measure of resolution (in the Rayleigh sense
> of an instrument's ability two resolve two points) but has been a measure
> of self-consistency. One can then argue about the statistics of this
> measure, but please keep in mind that this still does not address the
> problem that one is still using this to estimate what is reproducible in a
> reconstruction and what is noise. This becomes most obvious if this measure
> is applied to a structure where symmetry has been imposed, but I believe
> that it is perfectly general. If one imposes the wrong symmetry, one can
> still measure a self-consistency between two different reconstructions, but
> this has no reality in terms of actual resolution (Egelman, eLife, 2015).
> Regards,
> Ed
>
> On 8/12/15 8:51 AM, Frank, Joachim wrote:
>
> IT DOESN'T.
>
> Joachim.
>
>
> --
>
>
> Edward H. Egelman, Ph.D.
>
> Professor
>
> Dept. of Biochemistry and Molecular Genetics
>
> University of Virginia
>
>
>
> President
>
> Biophysical Society
>
>
>
> phone: 434-924-8210
>
> fax: 434-924-5069
>
> egelman at virginia.edu
>
> <http://www.people.virginia.edu/%7Eehe2n>
> http://www.people.virginia.edu/~ehe2n
>
>
> _______________________________________________
> 3dem mailing list
> 3dem at ncmir.ucsd.edu
> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
>
>


-- 
Dr. Jose-Maria Carazo
Biocomputing Unit, Head, CNB-CSIC
Spanish National Center for Biotechnology
Cell: +34639197980
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20150812/8bf6bf3b/attachment-0001.html>