Colony recognition and growth measurement for microbial imaging
Project description
Processing Images Easily (PIE) automatically tracks growing microcolonies in low-resolution brightfield and phase-contrast microscopy images. The program works for recognizing microcolonies in a wide range of microbes, and allows automated measurement of growth rate, lag times, and (if applicable) fluorescence across time for microcolonies founded by single cells. PIE recognizes colony outlines very robustly and accurately across a wide range of image brightnesses and focal depths, and allows simultaneous measurements of growth properties and fluorescence intensities with very high throughput (in our lab, ~100,000 colonies per experiment), including in complex, multiphase experiments.
To learn how to install and use PIE, see the PIE documentation.
To learn how PIE works, see our preprint.
To test microcolony recognition and growth tracking on your images, try our web application.
PIE Quickstart
Below is a quick reference to essential PIE functions; see the full PIE documentation for more details. If you have any questions about setting up or using PIE, we’d love to help! Feel free to contact us at pie-siegal-lab@nyu.edu or open an issue on github.
All the commands below must be run in Terminal (MacOS/Linux) / Command Prompt (Windows)
Installing PIE
PIE requires Python 3.6+, and can be installed using pip, which should come with your python installation.
In unix/macOS Terminal, run:
python -m pip install git+https://github.com/Siegallab/PIE@v1.0.1or, in Windows Command Prompt, run:
py -m pip install git+https://github.com/Siegallab/PIE@v1.0.1See Installing PIE for details.
Analyzing a single image
To run PIE on a single image:
pie analyze_single_image INPUT_IM_PATH OUTPUT_PATH IMAGE_TYPEWhere:
INPUT_IM_PATH is the path to the image you want to analyze
OUTPUT_PATH is a directory you’d like to store the results of the analysis in
IMAGE_TYPE is ‘bright’ (for cells that are brighter than the image background) or ‘dark’ (for cells that are darker than the image background)
See Running PIE single-image analysis for details on the inputs and outputs, as well as additional analysis options.
Analyzing timelapse experiments
To analyze a time lapse experiment, you need to create a setup file containing analysis parameters, and then run the analysis itself.
To interactively create a setup file:
pie run_setup_wizardTo analyze the timelapse experiment:
pie run_timelapse_analysis PATH_TO_SETUP_FILEWhere PATH_TO_SETUP_FILE is the path to the setup file created by the setup wizard.
See Running PIE timelapse experiments for information on analyzing complex, multi-phase experiments.
Creating movies
After timelapose experiments are analyzed, PIE can create movies of the output; this is helpful in understanding whether the analysis worked as expected.
To create simple movies of PIE analysis output for a single imaging position:
pie make_position_movie XY_POS PATH_TO_SETUP_FILEWhere:
XY_POS is the imaging position number for which the movie should be created (see Filenames for information on how to encode imaging position in filenames and the setup file)
PATH_TO_SETUP_FILE is the path to the setup file created by the setup wizard
See Creating movies of image analysis results for additional options and examples of more movie types that can be created from PIE output.
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
