Awesome

Monet

Note: This repository contains the scRNA-Seq analysis software. For other tools named Monet, see Disambiguation

Monet is an open-source Python package for analyzing and integrating scRNA-Seq data using PCA-based latent spaces. Datasets from the Monet paper (Wagner, 2020) can be found in a separate repository.

For questions and requests, please create an "issue" on GitHub. For a version history, see CHANGES.

Getting started

Installation

The recommended way to install Monet is to first install most of its dependencies using conda, and to then install Monet and other dependencies that are not available through conda using pip.

1. Installing Miniconda

If you are new to conda, please install Miniconda.

2. Create a new conda environment for installing Monet

Create a new conda environment named "monet" with Python 3.8 as follows (commands are for Linux/Ubuntu):

$ conda create -n monet python=3.8

3. Use conda to install most of Monet's dependencies

Activate the new environment and install the following packages:

$ conda activate monet
(monet) $ conda install scikit-learn pandas cython plotly seaborn statsmodels numba pytables networkx click

4. Use pip to install the remaining dependencies and Monet itself

Make sure your conda environment is still activated. Then install the following packages:

(monet) $ pip install leidenalg scanpy monet

Tutorials (v0.2.2)

The following tutorials were developed using Monet v0.2.2. They demonstrate how to use Monet to perform various basic and advanced analysis tasks. The Jupyter electronic notebooks can be downloaded from GitHub.

Basics

1. Loading and saving expression data
2. Importing/exporting data from/to Scanpy
3. Visualizing data with t-SNE

Clustering

1. Clustering data with Galapagos (t-SNE + DBSCAN)
2. Annotating clusters with cell types (coming soon)

Denoising

1. Denoising data with ENHANCE

Data integration

1. Training a Monet model (for integrative anlayses)
2. Plotting a batch-corrected t-SNE using mutual nearest neighbors (Haghverdi et al.%2C 2018)
3. Transferring labels between datasets using K-nearest neighbor classification

Copyright and License

Copyright (c) 2020-2021 Florian Wagner

Monet is licensed under an OSI-compliant 3-clause BSD license. For details, see LICENSE.

Disambiguation

The following other tools have been named Monet (styled either MONET or MONet):

• Overview of the Model and Observation Evaluation Toolkit (MONET) (Baker and Pan, 2017) [github]
• MONet: Unsupervised Scene Decomposition and Representation (Burgess et al., 2019) [github]
• MONET: a toolbox integrating top-performing methods for network modularization (Tomasoni et al., 2020) [preprint] [github]
• Multi-Objective Cellular Evolutionary Algorithm (MONET) (García-Nieto et al., 2019) [github]
• MONET: Multi-omic patient module detection by omic selection (Rappoport et al., 2020) [github]

Thanks to Michał Krassowski (@krassowski_m) and Dr. Matthias Stahl (@h_i_g_s_c_h) for providing these references.