[3dem] simulated micrograph for ctf determination?

[Fred Sigworth]

Making random projections, operating with CTF and adding noise is  
straightforward with any of several software packages (I can even give  
you a script in Matlab!).  But, there is a critical problem for which  
I haven't seen a solution.  This is the problem of solvating the  
macromolecule.  If you just take the pdb coordinates to create a  
density map, you will have a model of protein in a vacuum, which has a  
much higher-contrast molecular boundary than the actual molecule  
solvated in water.

Does anyone have a solution to this problem?  To get a realistic model  
of a streptavidin tetramer, we wound up doing a molecular dynamics  
simulation (L. Wang et al., J. Struct. Biol. 2008, PMID: 18707004).   
Using VMD Liguo Wang put the molecule into a box of water molecules,  
and then ran an MD simulation using NAMD for about a nanosecond to  
allow waters to find good positions.

-Fred Sigworth


> We are also interested in something similar. We searched but didn't  
> find programs that can take a known protein complex, for example  
> from the PDB, and automatically generate random noisy projections  
> similar to what could be a real single particle experiment. Ideally  
> we would like to be able to control image properties (level and type  
> of noise, CTF, etc.) so to parameterize and quantify our  
> classification methods accordingly.
>
> Any recommendations are really appreciated.
>
> - Alex.
>
> Steven Ludtke wrote:
>> There are several programs which will do this. For my part, you can  
>> use ctfit or applyctf  in EMAN1. However,
>> even with realistic pink noise added, these images still remain an  
>> imperfect simulation of real images. Different
>> programs use different strategies for the raw image processing  
>> (working with the whole micrograph, boxed out
>> particles, boxed out carbon film, etc.), so it's hard to come up  
>> with a single uniform test for them all...
>> ----------------------------------------------------------------------------
>> Steven Ludtke, Ph.D.
>> Associate Professor, Dept. of Biochemistry and Mol. Biol.       
>> Those who do
>> Co-Director National Center For Macromolecular Imaging            ARE
>> Baylor College of Medicine                                     The  
>> converse
>> sludtke at bcm.edu <mailto:sludtke at bcm.edu>  -or-  stevel at alumni.caltech.edu 
>>  <mailto:stevel at alumni.caltech.edu>               also applies
>> http://ncmi.bcm.edu/~stevel
>> On Sep 10, 2009, at 4:55 PM, jinghua tang wrote:
>>> To test various programs for determine the CTF, or simply the  
>>> defocus, does anyone have
>>> a way to generate a simulated micrograph in which different CTF  
>>> was applied?
>>>
>>> Thank you in advance for your kind helps!
>>>
>>> Jinghua
>>> _______________________________________________
>>> 3dem mailing list
>>> 3dem at ncmir.ucsd.edu <mailto:3dem at ncmir.ucsd.edu>
>>> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
>> ------------------------------------------------------------------------

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mail.ncmir.ucsd.edu/mailman/private/3dem/attachments/20090916/5e2d9a1d/attachment.html