[3dem] Spider-reference free classsum

[Edward Brignole]

Xiao,

Your spider script is looking for particle names with 5 digits (as in xyz00001), but in your email you say that your particle names have 4 digits (xyz0001).

If by "particle set" you mean that your particles were in a stack rather than as individual particles, then you need to indicate that by  "xyz@" instead of "xyz". One advantage of using a stack of particle images is that the number of digits doesn't matter, for instance: xyz at 00001 is the same as xyz at 0001.

One last comment, in your script the alignment radius set to 38 pixels won't work since you indicate that your particle dimension is 72 pixels and you are masking out the corners of your particles using a radius of 35. Since you set the diameter to 40 pixels you probably want to use 20 pixels for the last ring.

Good luck.
Ed

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Edward Brignole
Francisco Asturias Lab
The Scripps Research Institute
www.scripps.edu/~brignole

On Jan 31, 2010, at 7:54 PM, xiao.yang at studentmail.newcastle.edu.au wrote:

> Hi all,
> 
> I am new to Spider and would like to use it to make reference free 2d classsums from picked particle set (7850 images). When I did the run, the program was asking for the individual particles (ykZ0001) rather than a particle set. The batch-file (below) was modified from the nature paper. Can anyone be so kind to let me know which bit I was missing?
> 
> Thanks in advance
> 
> Xiao
> 
> 
> 
> ;-----------------------------------------------------------------------------!
> ; b03.fed/hbf : - iterative reference free alignment untilted-specimen images
> ;               - first part using AP SR reference free alignment :
> ;-----------------------------------------------------------------------------!
> 
> ; ----- Parameters -----
> X63=1.       ; first image number
> X62=7580      ; last image number
> x50=72      ; image dimension
> x51=x50/2+1  ; image center coordinates
> 
> ; ---- Input files ----
> fr l
> [untilted_images]../images/ykz    ; untilted windowed image filename template
> 
> ; ---- Output files ----
> fr l
> [masked_untilted]../images/mas    ; untilted masked images
> fr l
> [apsr_image]../r2d/avs***         ; AP SR output images
> fr l
> [apsr_doc]../doc/davs***          ; AP SR output doc files
> 
> ; -- END BATCH HEADER --
> 
> md
> tr on
> 
> ;-----------------------------------------------------------------------------!
> ;   Apply a circular mask with Gaussian edge on all images   
> ;-----------------------------------------------------------------------------!
> 
> ;   loop through particles
> DO LB1 x10=X63,X62
> 
> ;   apply a circular mask on the raw images and stack them 
>    ma
>    [untilted_images]{*****x10}
>    [masked_untilted]{*****x10}
>    (35.)    ; mask radius
>    g        ; Gaussian cutoff
>    c        ; background set to circumference average
>    x51,x51  ; mask center
>    (2.5)    ; Gaussian falloff
> LB1
> 
> ;-----------------------------------------------------------------------------!
> ;   Apply a reference free alignment as developped by 
> ;   Pawel Penczek et al., Ultramicroscopy 40 (1992) 33-53
> ;-----------------------------------------------------------------------------!
> 
> ap sr
> [masked_untilted]*****
> x63-x62        ; first and last image numbers
> (40)           ; expected size of the particle in pixels
> (5,38)         ; first and last rings used for rotationnal alignment
> *              ; image to use for centering (* = a gaussian disk is used)
> [apsr_image]   ; template for output global images
> [apsr_doc]     ; template for output doc files
> 
> ; delete intermediate files
> de a 
> [masked_untilted]00001
> 
> en
> 
> _______________________________________________
> 3dem mailing list
> 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/20100131/52ebefe8/attachment.html