Maximum Intensity Projections in Amira 5.4

(A) Setting Up Your Environment to Run Amira

1. Amira can currently only be run from jane, so you will first need to ssh to jane:

$ ssh -Y jane.crbs.ucsd.edu

2. Enter the directory that contains your reconstruction. For example:

$ cd /ccdbprod/ccdbprod2/home/CCDB_DATA_USER.portal/CCDB_DATA_USER/acquisition/project_20191/microscopy_87042/reconstruction 

3. Start Amira:

$ /ncmir/local.linux.amd64/Amira-5.4/bin/start

If you want, you can setup your login script to alias this command to the command 'amira'. To do this, add the following line to your .bashrc file by editing it in a text editor (vi, emacs, nano, etc.):

alias amira='/ncmir/local.linux.amd64/Amira-5.4/bin/start'

If you do this, the next time you use Amira, you will only have to type the command 'amira' rather than the full path.

(B) Preparing Volumes

Your reconstruction should already be in the MRC format if it was calculated using IMOD or TxBR. If your reconstruction is from a 4k x 4k camera, you should be able to load the full volume into memory on jane without any performance issues. However, if it is from the 8k x 8k camera or is a montage, you will need to bin it so Amira's performance will not suffer.

1.  If your volume needs to be binned, use the following command to bin by a factor of 2:

$ newstack -bin 2 basename.rec basename_2Dbin2.rec

2. When you perform the maximum intensity projection (MIP) calculations, you need to make sure your volume is devoid of all gold particles, as they will dramatically skew the calculation and result in unwanted artifacts. Open your original reconstruction file in imod. Find the minimum and maximum Z-slices at which the volume is free of gold particles. Make note of these two values, and subtract one from each of them (we'll call the final values Z_min and Z_max). We will create a second MRC file in which these Z-slices are trimmed out in the next step.

3. To perform the MIP calculations, you also need to have a volume with the contrast inverted such that electron dense regions appear white. This is opposite from what we are used to. Since the MIP calculations take a long time and the MIP rendering really slows down Amira, it is helpful to perform the calculations on a binned volume.

We can combine the binning, contrast inversion, and trimming all in one command.

If you have already binned your volume in Step 1, run the following command to invert the contrast:

$ newstack -scale 255,0 -secs Zmin-Zmax basename_2Dbin2.rec basename_2Dbin2_inverted.rec

If you haven't already binned your volume because it is from a 4k x 4k camera, run the following command to perform the binning and contrast inversion operations simultaneously:

$ newstack -bin 2 -scale 255,0 -secs Zmin-Zmax basename.rec basename_2Dbin2_inverted.rec

(C) Loading the Volumes into Amira

The Amira GUI is divided into four main frames. In the image below, I've given them names that I'll refer to for the rest of this tutorial:

1. In the Pool frame, click the green 'Open Data' button. Select your current directory from the upper pull-down menu (it should be the top choice), and select the first volume (basename.rec OR basename_2Dbin2.rec). Click the 'Load' button and select 'MRC Volume' in the 'Dialog' pop-up window. Click the 'OK' button.

2. After a small delay, you should see a new green button with your reconstruction name appear in the Pool frame.

3. Repeat step 1 for your volume with inverted contrast (basename_2Dbin2_inverted.rec). You should see a second green button appear in the Pool frame.

Pool --> Open Data --> Select Data --> MRC Volume

Right click on button --> Display --> Voltex

Left click on Voltex button

Color table

Luminance/alpha

Colormap --> Edit --> Adjust to histogram, then adjust the min and max values accordingly

Alpha scale = 1

Texture mode: 2D

Downsample: 2 2 2

Apply