PFT2 & EM3DR2 Tutorial

Steps to do before running PFT2 and EM3DR2


Before running pft2, you must complete the following steps:


1.  Scan micrograph images


2.  Box and extract individual particles


3.  Use a program such as the Bsoft program Bshow to fit the Contrast Transfer Function (CTF) in order to correct for the effects of defocusing.


4.     Combine all your data into a single project STAR file.

a.     Each individual data block (data for one micrograph) should start with the line data_ and define the parameters given in the example file.

b.     You can input the values via a text editor or you may use the Bsoft program bmg to define global parameters (i.e. parameters that will have the same value in each micrograph):  for example, to set pixel size run command "bmg -verbose 7 -Pixelsize 6.89 -output polyoma.star polyoma.star"

Example of initial project STAR file
Example of multiple micrograph STAR file (this example includes particle orientations and origins, which are not necessary for the initial STAR file)


5.     The following is a way to organize your data into subdirectories:
 
Directory
 File(s)
iterations
 location of STAR input and output files
particles
 picked particles
micrographs
 micrograph files
maps
 3D reconstructions

Download the following files to run a practice reconstruction:  polyoma_images.pif (particle images), polyoma_3d.pif (starting map).  These are images of murine polyomavirus from a reconstruction published in Ultramicroscopy 48:347-358 (1993):

Murine polyomavirus, surface rendering
Surface rendering of polyomavirus reconstruction

Note:  if you use the recommended subdirectory structure, then add "../directory_name/" in front of the particle and micrograph files in the example STAR file.


6.     In TEM, highest densities may either be black or white.  Images and starting map must be consistent.  The 3D-EM conventions state that the highest densities should be positive and are usually white; however, pft2 only requires that the contrast be consistent between particle images and maps.  Be sure the starting map and images have the same contrast.  To invert densities, use the Bsoft program bimg:

Example:  bimg -invert -verbose 3 polyoma_images.pif polyoma_images.pif

Note: if you have a lot of files you need to invert, make a script.


7.     Particle images and the starting map should have a pixel size within ~ 3% of each other.


8.     Use pft2 to find which radii show an acceptable correlation between the starting model and the particles.  Enter the command

pft2 -mode sphere_search -map polyoma_3D.pif -resolution 90,25 -verbose 2 polyoma.star

This command outputs an empft.rads file.  Plotting the radii against the correlation coefficients (use average data at end of the empft.rads file) will help you decide what the limits of the radial range should be, example plot for polyoma data.

Note:  to greatly speed up this step and other PFT2 runs, use the binning option, e.g. "-bin 2".



Click to continue to the reconstruction protocol

Return to main tutorial page

Return to main PFT3DR page

4 Jan 2008
Questions or problems?  Contact David_Belnap@byu.edu