Skip to content

Single Cell Outlier Selector - quickly find outliers in your single-cell data

License

Notifications You must be signed in to change notification settings

jfaccioni/SCOUTS

Repository files navigation

SCOUTS - Single Cell Outlier Selector

SCOUTS icon

SCOUTS is a tool that quickly finds OUTLIERS in single-cell datasets, especially those obtained by mass cytometry (CyToF) and single-cell RNA sequencing (scRNA-Seq) platforms.

Why SCOUTS?

Many single-cell analytical pipelines require some level of programming knowledge in order to be used. While some great tools for languages like R, Python and Julia have been developed, the entry-level barrier of programming is still intimidating for many scientists starting on the field of single-cell analysis. With this in mind, we developed SCOUTS to simplify this process. Through a desktop application, the user is able to choose the parameters for the outlier selection, and leave the hard work of programmatically subsetting the data to SCOUTS.

As a showcase of how to interpret and explore the data generated by SCOUTS, we also developed SCOUTS-violins, a secondary desktop application which displays the outlier populations identified by SCOUTS as violin plots.

SCOUTS basic workflow

Getting Started

Installation

SCOUTS is available as a:

  • Python package from PyPI - install with pip
  • Conda package - install with conda
  • GitHub repository - download/clone the repository
  • binary release (experimental)

Full installation instructions are found here.

Binary releases are found here. Keep in mind that these are experimental, and may not work properly on your system. In this case, we recommend following another installation method.

For any installation option (other than the binary release), SCOUTS requires Python >= 3.6.

Basic usage

The scouts package include two user interfaces:

SCOUTS: used to select outliers in a population of single-cells.

SCOUTS-violins (optional): used to visually inspect outliers selected by SCOUTS.

Troubleshooting

If you run into problems or errors during installation, please read our troubleshooting section.

Documentation

Read the full documentation here.

Citation

If you use SCOUTS, please cite us.

Authors

Juliano Faccioni - Programming and GUI development

Giovana Onzi - Concept and testing

License

This project is licensed under the MIT License.

Acknowledgements

Scientific counselling

  • Prof. Dr. Guido Lenz

Funding

  • CAPES/CNPq
  • NIH