IHC_DAB-Milton

Running DAB counts on Milton

What is everything

• bac_environment.yml - assist setting up conda environment so things work, may still need to install a few things with pip depending on underlying OS
• IHC_DAB_Count.py - the main analysis code
• IHC_Test.ipynb - Jupyter notebook that can help determine thresholds and make sure everything is working
• getTifs.py - Recursively search for TIF files in a base directory and save to a text file for input into submission script
• submit.sh - SBATCH submissions script to milton

First things first

This tool is for analysing images from the 3d-histec slide scanner that have been extracted to TIF. If they're still on casecenter or are mxrs files then I suggest tackling them with Qupath

Setting up environment

If you've not used conda on mitlon before, install it with:

module load anaconda3
conda init

Then create an environment from the yml conda create env -f bac_environment.yml

IHC_DAB_Count.py

Relatively simple analyis script that performs DAB scoring from command line. Simply takes a tif file location as input and writes out a number of txt files (one for each tile - script takes care of tiling)

python IHC_DAB_Count.py [pathToTif]

getTifs.py

Utilty to generate a list of images to pass to sbatch. Optionally give output filename, defaults to images.list

python getTifs.py [basePath] -o [outputFilename]

###submit.sh

See comments in file for things to change. Things to check:

• ntasks = the number of lines in images.list (or at least the amount you want to run)
• array = The range of lines you want to run on (ie, 1-10 or 1,4,7 etc)
• the correct file list is being used (images.list etc)