Skip to content

JinmiaoChenLab/scTM

Repository files navigation

scTM: A pacakge for topic modelling in transcriptomics data

image

Documentation Status

scTM is a package for spatial transcriptomics for single cell that uses topic modelling, solved with stochastic variational infernce. The interesting part is with the formulation of topic models, we can get interpretable embedding which are useful for downstream analysis.

Currently available modules: STAMP

Features

  • STAMP: A spatially-aware dimensional reduction designed for spatial data.

Minimal Installation

conda create --name sctm python=3.8
conda activate sctm
pip install sctm

or

conda create --name sctm python=3.8
git clone https://github.com/JinmiaoChenLab/scTM.git
conda activate sctm
cd scTM
pip install .

Basic Usage

Check out our usage of STAMP in the documentation with a simulated data at https://jinmiaochenlab.github.io/scTM/notebooks/stamp/Simulation.

The simulated data can be found at data/simulation.h5ad.

For more advanced usage please check the examples in our detailed tutorials.

Tested with

python == 3.8

numpy == 1.24.4

numba == 0.57.1

torch == 2.0.3

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

About

scTM: A package for topic modelling in transcriptomics data

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages