An unofficial, beginner-friendly collection of Colab notebooks for estimating cell potency from single-cell RNA-seq data using CytoTRACE2
This repository is intended for anyone, especially those with limited biology background who wants to explore CytoTRACE2, a tool that predicts how stem-like or differentiated a cell is using single-cell RNA-seq data.
These examples are based on the official CytoTRACE2 GitHub, with added explanations to help others like me who are new to this topic understand what's going on.
This project is a work in progress and will continue to improve as my understanding evolves.
CytoTRACE2 is a deep learning model that uses gene expression data to estimate a cell's potency, i.e. how likely it is to be a stem-like or undifferentiated cell. It's useful for studying:
- How cells differentiate over time
- Tumor stemness
- Developmental processes
| File/Folder | Description |
|---|---|
example_pancreas_colab.ipynb |
A notebook that runs CytoTRACE2 on a mouse pancreas dataset.  |
example_cordblood_colab.ipynb |
A notebook for applying CytoTRACE2 to human cord blood immune cells. |
immunogenicity_score_colab.ipynb |
(in preparation) An exploratory notebook on how CytoTRACE2 scores may relate to immune marker expression. |
requirements.txt |
Python dependencies to run the notebooks. |
git clone https://github.com/richcmwang/cytotrace2-for-beginners.git
cd cytotrace2-for-beginners