[3dem] [ccpem] Correlation coefficient em structures

[Dario Saczko-Brack]

Hi Eike,

as already mentioned, it depends how the density maps were created. If 
the two are from the same project or even within the same data set, 
SPIDER <https://spider.wadsworth.org/spider_doc/spider/docs/spider.html> 
offers a convenient way to calculate the cross correlation coefficients 
<https://spider.wadsworth.org/spider_doc/spider/docs/man/ccc.html>. You 
can also apply a mask to focus the ccc-calculation on specific areas.

Best


Dario


On 11.10.2017 16:05, Marin van Heel wrote:
>
> Hallo Eike,
>
> Steve is right... complicated topic!
>
> You may want to look at some of the original papers on this issue:
>
> 1) George Harauz and Marin van Heel, Exact filters for general 
> geometry three dimensional reconstruction, Optik 73 (1986) 146-156.
> 2) Marin van Heel and George Harauz, Resolution criteria for three 
> dimensional reconstructions, Optik 73 (1986) 119-122
> 3) Marin van Heel, Similarity measures between images, Ultramicroscopy 
> 21 (1987) 95-100.
> 4) Marin van Heel, Michael Schatz, and Elena Orlova, Correlation 
> functions revisited, Ultramicroscopy 46 (1992) 304-316.
> 5) Marin van Heel: Unveiling ribosomal structures: the final phases. 
> Current Opinions in Structural Biology 10 (2000) 259-264.
>
> My 5 cents...
>
> Marin.
>
> PS There are more pennies in my purse, and I can provide you with the 
> pdfs if needed.
>
> On 11/10/2017 10:28, Ludtke, Steven J wrote:
>> Hi Eike,
>> correlation coefficient is not normally a very useful metric for 
>> comparing two different maps, unless they happen to be generated in 
>> nearly identical ways. It can be useful for looking at self-symmetry 
>> and a few other things, but it is extremely sensitive to alignment, 
>> and perhaps even more importantly, to how the maps have been 
>> filtered.  Normally if you wish to compare two maps you would do it 
>> using an FSC, which is resolution dependent and not sensitive to 
>> filtration differences. While this will give you a curve instead of a 
>> single number, it is much more useful in assessing how two maps 
>> differ. While there are many different tools you can use to generate 
>> one, this is EMAN2's method:
>>
>> e2proc3d.py map1.mrc fsccurve.txt --calcfsc map2.mrc # any input file 
>> format is fine
>>
>> To see the results:
>> e2display.py --plot fsccurve.txt
>>
>> An alternative which can compute local FSC, and also produce a 
>> locally filtered average map based on the local resolution:
>> e2fsc.py map1.mrc map2.mrc --output fscmap.hdf --outfilt localfiltmap.mrc
>>
>> If the maps are not properly aligned, you can do something like 
>> before computing the FSC:
>> e2proc3d.py map1.mrc map1_aligned.mrc 
>> --align rotate_translate_3d_tree --alignref map2.mrc
>>
>>
>>> On Oct 11, 2017, at 7:08 AM, Schulz, Eike-Christian 
>>> xxxxxxxxxxxxxxxxxxxxxxx wrote:
>>>
>>> ****CAUTION:*** This email is not from a BCM Source. Only click 
>>> links or open attachments you know are safe.*
>>> ------------------------------------------------------------------------
>>>
>>> Dear all,
>>>
>>> Is there a way to calculate structural similarity of electron 
>>> density maps in electron microscopy? I am looking for a solution 
>>> that would report e.g. a correlation coefficient …
>>>
>>> With best regards,
>>>
>>> Eike
>>>
>>
>> --------------------------------------------------------------------------------------
>> Steven Ludtke, Ph.D.
>> Charles C. Bell, Jr. Professor of Biochemistry and Mol. Biol.
>>
>
> -- 
> ==============================================================
>
>      Prof Dr Ir Marin van Heel
>
>      Laboratório Nacional de Nanotecnologia - LNNano
>      CNPEM/LNNano, Campinas, Brazil
>
>      tel:    +55-19-3518-2316
>      mobile  +55-19-981809332
>                   (041-19-981809332 TIM)
>      Skype:  Marin.van.Heel
>      email:  marin.vanheel(A_T)gmail.com
>              marin.vanheel(A_T)lnnano.cnpem.br
>      and:    mvh.office(A_T)gmail.com
>
> --------------------------------------------------
>      Emeritus Professor of Cryo-EM Data Processing
>      Leiden University
>      Mobile NL: +31(0)652736618
>
>      Emeritus Professor of Structural Biology
>      Imperial College London
>      Faculty of Natural Sciences
>      email: m.vanheel(A_T)imperial.ac.uk
>
> --------------------------------------------------
>
> I receive many emails per day and, although I try,
> there is no guarantee that I will actually read each incoming email.
>
>
> _______________________________________________
> 3dem mailing list
> 3dem at ncmir.ucsd.edu
> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20171011/f5fca0c9/attachment-0001.html>