From: Francis E Reyes
Date: 11 October 2011 19:21
http://hoowstuffworks.blogspot.com/2011/10/adobe-demos-amazing-unblur-feature.html
Though I can't really see the image myself... the gasp of the audience is telling
With respect to existing density modification programs, I wonder if such technology (whatever it is) can ever clear up my messy density maps.
F
---------------------------------------------
Francis E. Reyes M.Sc.
215 UCB
University of Colorado at Boulder
----------
From: Pete Meyer
I could be wrong, but my understanding is that they're removing motion blur from the image - so I don't think it'll be directly applicable to density modification.
But I'd be very happy to be wrong on this one.
Pete
----------
From: Ethan Merritt
The technology is directly parallel to TLSMD analysis.
It creates a model that includes a prediction of the blurring in the
original image, or in our case the blurring in the map generated by
Fourier transform of the original data. The difference is that in the
case of a photograph one is interested in the altered image rather than
the underlying model. In the case of a crystal structure one is interested
in the underlying model rather than the electron density map per se.
I have for a long time used the "motion-blurred photograph" parallel
to explain how TLSMD works. See for example page 3 of
http://www.ccp4.ac.uk/schools/APS-2009/tutorials/tls/Merritt30th.pdf
Ethan
--
Ethan A Merritt
Biomolecular Structure Center, K-428 Health Sciences Bldg
University of Washington, Seattle 98195-7742
----------
From: Garib N Murshudov
Date: 11 October 2011 19:21
http://hoowstuffworks.blogspot.com/2011/10/adobe-demos-amazing-unblur-feature.html
Though I can't really see the image myself... the gasp of the audience is telling
With respect to existing density modification programs, I wonder if such technology (whatever it is) can ever clear up my messy density maps.
F
---------------------------------------------
Francis E. Reyes M.Sc.
215 UCB
University of Colorado at Boulder
----------
From: Pete Meyer
I could be wrong, but my understanding is that they're removing motion blur from the image - so I don't think it'll be directly applicable to density modification.
But I'd be very happy to be wrong on this one.
Pete
----------
From: Ethan Merritt
The technology is directly parallel to TLSMD analysis.
It creates a model that includes a prediction of the blurring in the
original image, or in our case the blurring in the map generated by
Fourier transform of the original data. The difference is that in the
case of a photograph one is interested in the altered image rather than
the underlying model. In the case of a crystal structure one is interested
in the underlying model rather than the electron density map per se.
I have for a long time used the "motion-blurred photograph" parallel
to explain how TLSMD works. See for example page 3 of
http://www.ccp4.ac.uk/schools/APS-2009/tutorials/tls/Merritt30th.pdf
Ethan
--
Ethan A Merritt
Biomolecular Structure Center, K-428 Health Sciences Bldg
University of Washington, Seattle 98195-7742
----------
From: Garib N Murshudov
Position (general motion blur is a special case of it) dependent blurring can be applied to denisty improvement but problem is extremely ill posed. While deblurring you need to reduce noise amplification. Proper regularisation needs to be designed, probem becomes NxN linear equation where N is the number of grid points in density which might be a little bit large. However iterative method exists in inverse problems field.
regards
Garib
Garib N Murshudov
Structural Studies Division
Structural Studies Division
MRC Laboratory of Molecular Biology
Hills Road
Cambridge
CB2 0QH UK
No comments:
Post a Comment