Exploring the latent space of transcriptomic data with topic modeling

See analyses for a detailed description of all the steps to reproduce the analyses

Environment

The easiest way to run all the notebooks of this repository with all dependencies handled correctly is using Docker.

You can enable jupyter lab with docker-compose

docker-compose up -d

then point your browser to localhost

A Python package with used functions can be downloaded from https://pypi.org/project/topicpy/

pip install topicpy

Name		Name	Last commit message	Last commit date
Latest commit History 192 Commits
.github		.github
colab		colab
datasets		datasets
hSBM_Topicmodel @ 886ee49		hSBM_Topicmodel @ 886ee49
tree_plotter @ b3a2f7c		tree_plotter @ b3a2f7c
unified		unified
.dvcignore		.dvcignore
.gitignore		.gitignore
.gitmodules		.gitmodules
CITATION.cff		CITATION.cff
Dockerfile		Dockerfile
LICENSE		LICENSE
README.md		README.md
WGCNA.ipynb		WGCNA.ipynb
alternative_clustering.ipynb		alternative_clustering.ipynb
analyses.md		analyses.md
bipartite_network.png		bipartite_network.png
docker-compose.yml		docker-compose.yml
hSBM_clusters.ipynb		hSBM_clusters.ipynb
hSBM_topic-dist.ipynb		hSBM_topic-dist.ipynb
hSBM_topics.ipynb		hSBM_topics.ipynb
hSBM_word-dist.ipynb		hSBM_word-dist.ipynb
install_pkgs.R		install_pkgs.R
latent_distances.ipynb		latent_distances.ipynb
predictor.ipynb		predictor.ipynb
requirements.txt		requirements.txt
run_graph.ipynb		run_graph.ipynb
single_cell.ipynb		single_cell.ipynb