Timelapse Feature Explorer is a web tool for interacting with and visualizing features in time-series segmented data. You can apply color maps and ranges, switch between features in your dataset, play through your data to observe motion, and view plots showing how feature data change over time.
This project originated from the Allen Institute for Cell Science (AICS) Nuclear Morphogenesis project and is being updated to support broader use cases. View our Issues page for more details about potential future features!
Stable build: timelapse.allencell.org
Latest (main branch): https://allen-cell-animated.github.io/timelapse-colorizer/
Datasets must be preprocessed for viewing in Timelapse Feature Explorer. We provide the colorizer-data Python package, which includes tutorials and documentation about the data specification. See our getting started guide for more details.
Compatible datasets hosted in an HTTPS-accessible location (ex: AWS S3 bucket, Google Cloud storage, etc.) can be loaded directly from the interface with just a URL. Click the Load button in the top right to open a URL.
For local datasets, you can install and use the tfe-open CLI tool included in the colorizer-data Python package. Please check the colorizer-data GitHub repository for the latest version.
# In a Python environment:
python -m pip install git+https://github.com/allen-cell-animated/colorizer-data.git@v1.7.0
tfe-open "/path/to/your/dataset"Installation is optional and is only necessary for accessing datasets via HTTP. You can use the hosted version of Timelapse Feature Explorer to access HTTPS-hosted datasets without installing the project.
Prerequisites:
- Node 18 or higher: https://nodejs.org
- Python 3 (and optionally, a virtual Python environment)
Open a command terminal and run the following commands. Installation may take a few minutes.
git clone https://github.com/allen-cell-animated/timelapse-colorizer.git
cd timelapse-colorizer
npm installTo start the development server, run:
npm run devThis will allow you to access the viewer from your browser. By default, the server will be hosted at http://localhost:5173/.
You can also access our internal build (which may have some additional experimental features) by running npm run dev-internal.
See CONTRIBUTING.md for information related to developing the code.

