snp2cell

snp2cell is a package for identifying gene regulation involved in specific traits and cell types. It combines three elements: (i) GWAS summary statistics, (ii) single cell data and (iii) a base gene regulatory network.

A network propagation approach is used to integrate and overlap different types of scores on the network. Random permutations of scores are then used to evaluate the significance of high scores.

As an output, a networkx graph of the gene regulatory network with integrated scores can be used to inspect gene regulatory programs that are linked to the trait (from GWAS) on a per cell type basis.

<img src="docs/assets/snp2cell_schematic.png" width="300">

System requirements

<details> <summary><b>show requirements</b></summary>

Hardware requirements

snp2cell can run on a standard computer with enough RAM to hold the used datasets in memory. It can make use of multiple CPUs to speed up computations.

Software requirements

OS requirements

The package has been tested on:

• macOS Monterey (12.6.7)
• Linux: Ubuntu 22.04 jammy

Python requirements

A python version >=3.5 and <3.12 is required for all dependencies to work. Various python libraries are used, listed in setup.py, including the python scientific stack, networkx and scanpy. snp2cell and all dependencies can be installed via pip (see below).

</details>

Installation

Optional: create and activate a new conda environment (with python<3.12):

mamba create -n snp2cell "python<3.12"
mamba activate snp2cell

from github

To download via ssh, an ssh key must have be added to your GitHub profile.

pip install git+ssh://git@github.com/Teichlab/snp2cell.git

Alternatively, clone the repository and install with pip in two steps:

git clone <preferred way ...>
cd snp2cell
pip install .

(installation time: around 2 min)

Usage

Python module

snp2cell can be imported as a python module (see notebooks for examples).

Demo: A minimal demo can be found here as a jupyter notebook and as a unit test in test/test_toy_example.py. (running time: around 12 sec)

CLI

Importing snp2cell as a python module gives most flexibility. Additionally, there is a command line interface. To see all options, run:

snp2cell --help

Optionally, activate autocompletion for the command line tool. E.g. for bash run:

snp2cell --install-completion bash
. ~/.bashrc

Example notebooks

• toy example
• GWAS analysis

Citation

snp2cell is part of the paper "A multi-omic atlas of human embryonic skeletal development" by To, Fei, Pett et al.; 2024; Nature (https://www.nature.com/articles/s41586-024-08189-z)