Update:
- This repo is designed to extract statistical information from raw Imaris file with (.ims) extension.
- This script does not calculate any statistical information, it simply extracts existing information from each .ims file and formats in a manner that helps with downsteam stream analysis software such as FlowJo.
- Tested on: Windows 10, Ubuntu 20.04
- Install Time: <10 minutes Create Enviroment:
conda env create -f environment.yml
- We currently offer 4 types of data parsers
- Surface Stats
- Surface Track Stats
- Surfaces at Unique Time Steps
- All Tracks
- Each of the jupyter notebooks for the parsers above uses a config file from parsers/config/ folder.
- To extract information from your own Imaris Files, simply grab the paths to each directory that contains the .ims files and update the config.yaml file, provide a save directory and run any one of the notebooks to extract the relevant information.
- For each file a formatted csv file will be generated that contains all relevant statistical information from each .ims file for further analysis.
- Parsers
- data
- parsers
- config
- imaris
- parsers
- scripts
- utils
- Shehan Perera
- Juraj Kabat