Differences

This shows you the differences between two versions of the page.

--- eman1:faq:multipleboxer [2025/07/05 15:24] – created steveludtke
+++ eman1:faq:multipleboxer [2025/07/05 15:26] (current) – steveludtke
@@ Line 1: / Line 1: @@
 **Q:** Boxer - I used a reference and batchboxer to pick particles from cryo-micrographs using a script. I did not use the minidist option to remove touching particles.
 **Q:** Boxer - I used the GUI boxer and the .box files to delete selected particles on the carbon file. I saved these new .box files
 **Q:** Boxer - How to I use the minidist option to remove particles in close contact without going through picking and manually removing particles again.
 ----
-CategoryCategory
+**A:**
+This is best done using a shell script. You can get boxer to preload a micrograph and a box database as it starts, for example:
+<code>boxer image.mrc box=image.box bright=0 contrast=0.2</code>
+So, you just write a shell script which calls boxer sequentially with each pair of image files and box databases. After saving the results within
+boxer, just quit, and the shell script will load the next image. Since boxer remembers window positions, everything should pop up pretty much like
+you left it.
+Note that boxer (and most other EMAN programs) append new images to existing files. That is, if you already have a file called img123.hed, and select
+this as a file to write particles to in boxer, it will append the new particles to the existing file rather than create a new one. For this reason, my
+approach is to:
+  - Run ''batchboxer'' on all images (using a shell script), but only output the box database, not the selected particles
+  - Run ''boxer'' on all images (using a shell script), again, only save the box database (this will be overwritten, not appended)
+  - When all of my box databases are good, then run ''batchboxer'' again to extract the particles, but don't have it autobox, just have it use the existing box database files.
+My own shell script (I use zsh) for the second step above looks like this (assumes .mrc extension on the micrographs):
+<code>foreach i (*box)
+boxer ${i%.box}.mrc box=$i bright=0 contrast=.15
+end</code>