Version 1.5.4

Picking 3D volumes from an image

Go here for a more general reference with pictures to picking particles in bshow.

Open a 3D image in bshow and before doing anything, make sure the pixel size is set correctly. Find an appropriate scaling and contrast to see the particles, perhaps using the "Averaging mode" to improve the visual effect.

Select the "Micrograph/Pick particles" menu item. Choose a box radius that would comfortably include a particle in all 3 dimensions. The bad area radius should typically be about half the box radius, but that is up to the user. For each particle, go through the slices to find the middle of the particle and click on the center to pick it. This 3D box can be moved in x and y to position it better. To reposition it in z, go up or down through the slices and click in a slice deemed to be cutting through the middle of the particle. If there are bad features close to a particle (i.e., intruding into the box), a bad area can be selected by the middle button (or control-left button) and positioned to cover an undesirable feature. Both boxes and bad areas can be deleted with shift-left button.

When all the particles have been picked, make sure the "Particles" window is in front by clicking on it. Then select the "Boxes/Extract particles" menu item and save it to a multi-image file (only the PIF, .pif, and SPIDER, .spi, formats are true 3D multi-image formats). This image file can be read with bshow to see if it looks acceptable.

Next click on the main window and select the "Micrograph/Write parameters" item to save the micrograph parameters to a STAR file (extension .star).  This parameter file can be read back into bshow with the associated micrograph or particle image file.

Aligning 3D volumes to a reference

Currently Bsoft offers only reference-based alignment. References are generally derived from atomic coordinates or other 3D reconstruction techniques (such as traditional SPA). Here it is important to get the sampling correct and consistently encode it into the image headers. The first task is to prepare the reference to the same pixel size and image size as the particles in the extracted volumes, and to orient it correctly according to its symmetry. This is a topic for another tutorial and I'll assume this has already been done with the reference.

The program bfind uses a brute-force cross-correlation search through the asymmetric unit of the reference to find the orientation of an extracted particle. The cross-correlation can be used with a reciprocal space mask, to eliminate the contribution of missing data, such as the missing wedge in the case of tomography. To create such a mask with a size of {70,70,70}, a wedge defined by the tilt angles {-60,60}, and wrapped around for reciprocal space:
bmissing -v 7 -wedge 35,-60,60 -create 70,70,70 -wrap mask.map

Next  determine the orientation of each particle:

bfind -verbose 1 -symmetry D5 -Template ref.map -angle 3 -resolution 40,200 -output particle_10.star -mgid MG1024 -image 9 -Mask mask.map -Image particle.pif particle_10.pif

The symmetry is important to limit the search to the asymmetric unit. The template is the reference map used. The angle option gives the angular step size for all 3 Euler angles, with phi and theta calculated to cover the asymmetric unit evenly. The resolution limits (units in angstroms) are imposed to avoid high resolution noise and low resolution features from dominating the cross-correlation. Only two 3D images are used, and the "-image" option specifies the image from a multi-image file to be used (image numbering starts at 0). The "-mgid" option sets the string for the micrograph identifer in the output STAR file. This must be set correctly to be able to reintegrate this STAR file with the STAR file written after picking the particles:

bmg -v 7 -input particle.star -output particle_new.star particle_10.star

The correctly oriented particles can be averaged to get a final reconstruction:

badd -v 7 -output particle_rec.pif particle_*.pif