[chimerax-users] Interactive map sharpening and filtering?

[Oliver Clarke]

There will definitely be interest I can guarantee it!

Any options for interactively filtering and sharpening are extremely helpful for anyone doing EM, and provide significant advantages over doing so in programs without GUIs (or which may not have GUIs on the server where they may be located).

Cheers
Oli
> On May 10, 2017, at 5:39 PM, Tom Goddard <goddard at sonic.net> wrote:
> 
> I guess by low-pass filtering you mean that in Fourier space the radial falloff scale factor is exactly 1.0 out to some frequency and then drops off gradually, where as a Gaussian convolution starts dropping from 1.0 gradually right from 0 frequency.  Using different kernels (radial scale factor functions) would be easy — they just need to be defined by parameters that can be passed to the command.  For low pass you probably need a minimum frequency and a maximum frequency giving the range of the rolloff.
> 
> We didn’t provide many map filtering options in Chimera 1, mostly you would use other software like EMAN2 to do that.  But I can extend the capabilities in ChimeraX if there is interest.
> 
> 	Tom
> 
> 
>> On May 10, 2017, at 4:29 AM, Aaron Lewis <alewis at mrc-lmb.cam.ac.uk <mailto:alewis at mrc-lmb.cam.ac.uk>> wrote:
>> 
>> I'd add that there is an important difference between Gaussian filtering and band-pass filtering: Gaussian filtering blurs (sharpens) everything by a specified amount, whereas low-pass (high-pass) filtering imposes a minimum amount of blurriness (sharpness). This often makes band-pass filtering the right choice for images with heterogenous resolution, because you often want to blur the noisiest, sharpest parts without blurring the smoothest, blurriest parts. Cryo-EM maps often fall in this category, and thus represent a common use case.
>> 
>> On Wed, May 10, 2017 at 2:54 AM Tom Goddard <goddard at sonic.net <mailto:goddard at sonic.net>> wrote:
>> This is just like ChimeraX Gaussian filtering which does a convolution with a Gaussian in Fourier space.  That is relatively fast — a few seconds for a 256**3 map.  You can already do sharpening with the “invert” option to the gaussian filtering command (“vol gaussian #1 1.5 invert true”).  Gaussian filtering is one form of low pass filtering — others could be added.  I think it would be nice to allow a very fast (10 updates per second) test filtering on a small region using a mouse drag to figure out the desired parameters.  But I never spent to the time to figure out a user interface (what small region, how do you undo, how do you then apply to full map, …).
>> 
>>         Tom
>> 
>> 
>> > On May 6, 2017, at 10:27 AM, Oliver Clarke <olibclarke at gmail.com <mailto:olibclarke at gmail.com>> wrote:
>> >
>> > Hi,
>> >
>> > Has any consideration been given to adding the capacity for fourier-space map sharpening/low pass filtering to ChimeraX? I’m not sure whether or not it is feasible speed wise, but if it is it would be a very useful addition for EM, allowing one to easily try different sharpening B-factors/low pass cutoffs and save the resulting map(s).
>> >
>> > Cheers
>> > Oli
>> > _______________________________________________
>> > ChimeraX-users mailing list
>> > ChimeraX-users at cgl.ucsf.edu <mailto:ChimeraX-users at cgl.ucsf.edu>
>> > Manage subscription:
>> > http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users <http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users>
>> >
>> 
>> 
>> _______________________________________________
>> ChimeraX-users mailing list
>> ChimeraX-users at cgl.ucsf.edu <mailto:ChimeraX-users at cgl.ucsf.edu>
>> Manage subscription:
>> http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users <http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users>
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimerax-users/attachments/20170510/4115188a/attachment.html>