Microspheres

Python and ImageJ/Fiji scripts to capture blockface images, process images, and consolidate data for microsphere experiments. Registration of blockface images to the Allen Mouse Brain atlas is completed using the QuickNII tool (https://www.nitrc.org/projects/quicknii) and the registration of microspheres to defined brain structures is done using Nutil (https://www.nitrc.org/projects/nutil).

General Info

To conduct our experiment we used a Raspberry Pi to do our imaging and Ubuntu 16.04LTS to do our data processing. The instructions for the setup of our hardware can be found below under Assembly, and the instructions for running our custom Python script for imaging can be found below under get_image.py instructions.

After individual brains are processed using Nutil, we utilize a custom Python data processing script to map the microsphere locations in each mouses' brain to different levels of the Allen Brain Atlas. The general usage of this custom script is explained below under localization.py instructions.

NOTES: Our scripts run on Python3 on Ubuntu/Windows 10. You need to have Pandas and Numpy in your environment and you may have to pip install an additional backend dependency to parse xlsx files such as xlrd or openpyxl.

When you are running $ python, please make sure that $ which python, points to your python3 installation (this command only works on Linux). If running $ which python points to /usr/bin/python27 or another path suffixed with a number containing "2", please try running the scripts with $ python3.

The dependencies for these scripts are very stable and different versions of the dependencies should all work, within reason.

Hardware Assembly

Blockface imaging was performed with a Raspberry Pi Camera (8MP, Camera Board V2) equipped with a varifocal (6-22mm) CCTV lens (Binchil, model #149129). An orange gelatin filter was cut to fit on the back aperture of the lens, and a blue LED ring light (Adafruit, New York, USA; Flora NeoPixel Ring, P1463) was mounted below the lens on a custom 3D printed camera mount (attached to Vibratome). Images were acquired on a Raspberry Pi 3 microcomputer (Raspberry Pi Foundation, Cambridge, UK) using get_image.py.

Cloning the Repo

1. make sure git is installed
2. press ctrl+alt+t to open a terminal window
3. $ cd ~
4. $ git clone https://github.com/silasilab/microspheres.git

get_image.py Instructions

This script takes an image with the CSI camera attached to the Raspberry Pi on user input.

1. press ctrl+alt+t to open a terminal window
2. $ cd ~
3. $ cd microspheres
4. inspect the imports of this script and pip install whatever you are missing
5. make a folder on your desktop called pi_brain (you can call it something else or use a different path, but be sure to change the savepath in get_image.py)
6. $ python get_image.py
7. to take an image press spacebar

BlockfaceImagingWorkflow.eNeuro.ijm

This script facilitates the processing of blockface images using ImageJ/Fiji

localization.py Instructions

Following microsphere localization with Nutil, this script takes Nutil input data and outputs an xslx spreadsheet containing a worksheet for each region level. Each row of each worksheet represents one mouse, and the columns represent a specific region. Spreadsheet[worksheet (level num)][mouse num][region num] == the count of microspheres.

1. press ctrl+alt+t to open a terminal window
2. $ cd ~
3. $ cd microspheres
4. open up localization.py in a text editor and change the variable named "base_dir" at the top of the file to point to where your input directory is. You can do this with $ vim localization.py, $ nano localization.py, or with whichever other text editor you like using.
5. $ python localization.py
6. inspect the output files