MDSINE2_Paper

Pull this code via

git clone --recurse-submodules https://github.com/gerberlab/MDSINE2_Paper.git

Important links

Associated GitHub repo for the Bayesian model: "MDSINE2"
Raw sequences from longitudinal experiments on NCBI:
Zenodo repositories containing pre-computed outputs of inference of the full dataset and cross-validation:
(CV + Replicates)
(Full Inference)
(Note: please ensure downloads are from the latest versions of these records. To extract a .tar.xf archive on Linux, use the tar xf command; on Windows, use 7-zip.)

Jupyter Notebooks

Folder containing tutorials as notebooks exploring the model, data and paper that can be opened directly in Google Colab:
Folder containing notebooks to reproduce figures found in the paper (requires local Jupyter instance to re-run)

Instructions

For details about individual steps in the analysis, refer to the primary internal README: scripts/README.md

As noted in that document, all scripts should be run from the scripts directory. This requirement is in place to match relative pathing of the root scripts/settings.sh environment file, from which all other paths are defined relatively.

Setup for analysis on a local machine

This section outlines how to run MDSINE2 analysis on our dataset in full, with bash, conda and git.

One must first install the MDSINE2 package. The recommended setup uses the conda recipe provided in that repository.

cd ~/work
git clone https://github.com/gerberlab/MDSINE2
cd MDSINE2
conda env create -f conda_recipe.yml 
conda activate mdsine2

Next, clone this repository which contains the data and scripts to perform the analysis.

cd ~/work
git clone https://github.com/gerberlab/MDSINE2_Paper
cd MDSINE2_Paper

Once the above installation is done, one should also install Jupyter/Jupyterlab for data visualization.

conda install -c conda-forge jupyter jupyterlab ipywidgets nodejs
jupyter nbextension enable --py widgetsnbextension

Analysis on a local machine

We provided shell scripts that implement several pipelines. Link: Analysis Pipeline

To reproduce the figures using the output of these analyses, please go through the Jupyter notebooks located in the paper_figures directory of this repository. To run them yourself, you will need to run a jupyter instance, for example:

cd MDSINE2_Paper
jupyter lab --port 8888

and navigate to paper_figures/ through JupyterLab.