**Multiscale hyperdimensional phenotypic analysis of organoids
Please check doc/index.html for more detailed documentation on how to get started with SCOUT. Thank you.
SCOUT has been tested estensively on a variety of Linux workstations with the following minimum specifications:
- 32 GB RAM
- 1 TB HDD
- 4-core i7 CPU
- Ubuntu 16.04 LTS or 18.04 LTS
Windows and Mac users may be able to install from source, but this has not been tested. On these platforms, Docker may be an easier option for installing and running SCOUT.
SCOUT was tested on Python 3.6 and 3.7. All of the required dependencies are listed in the requirements.txt file, and additional developement dependencies can be found in the environment.yml file for recreating a conda environment. Specific versions that were used for running SCOUT are also described in the specific_versions.txt file.
A guide for installing SCOUT is available here. Typically, takes ~15 minutes to install SCOUT on a Linux desktop computer.
The SCOUT documentation has a "Tutorial" section to show users how to use SCOUT. After installation, the tutorial is acts as a pipeline walkthrough for a typical two-group organoid analysis (treatment vs control). The expected runtime for the SCOUT pipeline on a single organoid is ~4 hours, and the overall runtime on a comparative study will scale with the number of organoids.
The notebooks/ folder contains Jupyter notebooks with all of the custom code
used to generate visualzations for presenting the main findings in the SCOUT
paper. After installing SCOUT, running jupyter notebook within the
notebooks/ folder will allow you to read and run these notebooks.