hormone2cell 0.1.1

A toolkit for predicting hormone producing and receiving strength in single cell datasets.

hormone2cell

A toolkit for predicting hormone producing and receiving strength in single cell datasets.

hormone2cell identifies hormone producing cell types (HPCs) and hormone receiving cell types (HRCs) from single-cell or single-nucleus expression data using a curated Hormone Receptor Reference Database. It applies cell-type-level expression filtering, multi-gene logic, and strength scoring to summarize predicted hormone production and reception across cell types.

Overview of the hormone2cell workflow, created with AI assistance.

Installation

We recommend installing hormone2cell in a clean conda environment.

Install from PyPI

# Create and activate conda environment
conda create -n hormone2cell_env python=3.10 -y
conda activate hormone2cell_env

# Install hormone2cell from PyPI
pip install hormone2cell

Install from GitHub for development

If you want to install the latest development version from GitHub:

# Create and activate conda environment
conda create -n hormone2cell_env python=3.10 -y
conda activate hormone2cell_env

# Clone repository
git clone https://github.com/Teichlab/hormone2cell.git
cd hormone2cell

# Install in editable mode
pip install -e .

Alternatively, download the package via Code → Download ZIP from GitHub and install it locally:

cd hormone2cell
pip install -e .

Usage and Documentation

Function docstrings and package documentation are available on the ReadTheDocs page.

The detailed tutorial is available as a notebook and as a rendered ReadTheDocs page.

The Hormone-Receptor Reference Database, including gene definitions for each hormone and receptor, is available from the Hormone Cell Atlas download page.

Other resources

For hormone-centric analyses and visualization, please visit the Hormone Cell Atlas website.

For the single-cell resource associated with this project, please visit the Hormone Cell Atlas single-cell resource.

Citation

Citation information will be added once available.

Parts of the method were informed by a previously published adaptive thresholding framework: Nature Communications, 2024.