Bayesian Survival Analysis of Electroporation Times from Molecular Dynamics

---

Authors: Sergio Pérez-Conesa, Lea Rems and Lucie Delemotte.

This project uses Bayesian Survival Analysis (PYMC3) to study the rate of membrane electroporation events. The first electroporation times are obtained from coarse-grained molecular dynamics simulations using realistic membrane compositions with a constant external electric field. The simulations were run by prof. Lea Rems and the bayesian inference by myself, Dr. Sergio Pérez-Conesa. We are both members of the Delemottelab led by prof. Lucie Delemotte. All the explanations can be found in the article and the rest of code and data here

I am happy to connect and discuss this and other projects through github, linkedin, twitter, email etc. Feel free to suggest ways we could have improved this code.

If you want to cite this code, please use CITE.bib, thank you!

Published Preprint: Identification of electroporation sites in the complex lipid organization of the plasma membrane

Published Article: Identification of electroporation sites in the complex lipid organization of the plasma membrane

Running the code

The code is based on the jupyter notebooks. is used to convert the raw data into. Notebooks `` do the actual calculations.

Binder

Try out my binder to run the notebooks live on the cloud! It's free 😉

Unfortunatelly, you can't train the models (too few memory on binder), but you can load them from the OSF repostory and analyze them in my binder.

Recreate conda environment

To recreate the conda environment used:

conda env create -f environment.yml

Use environment_exact.yml for the exact environment.

Getting additional data files

All the data, including the inference models, simulations etc. can be found in Open Software Foundation.

Project Organization

├── LICENSE
│
├── Makefile           <- Makefile with commands like `make update_data` or `make format`
│
├── README.md          <- The top-level README for developers using this project.
├── data
│   ├── external       <- Data from third party sources.
│   ├── interim        <- Intermediate data that has been transformed.
│   └── processed      <- The final, canonical data sets for modeling.
│
├── models             <- Learned models
│
├── notebooks          <- Jupyter notebooks.
│
├── reports            <- Generated analysis as HTML, PDF, LaTeX, etc.
│   ├── final_figures  <- Generated graphics and figures to be used in final report.
│   └── docs           <- Generated LaTeX, pdf, word, powerpoint etc.
│
├── environment.yml    <- The necessary packages to install in conda environment.
│
├── environment_exact.yml   <- The exact package versions used.
│
├── src                <- Source code for use in this project.
│   ├── __init__.py    <- Makes src a Python module
│   │
│   └── data           <- Scripts to download or generate data
│
└──

---

Project based on the cookiecutter for Molecular Dynamics. Which is itself based on the cookiecutter data science project template #cookiecutterdatascience

---

To Do

• Clean-up the repository of unused or WIP files.
• Make binder-friendly the notebooks.
• Make a version specific environment_exact.yml.
• Remove references to external scripts.
• Can we reproduce the data in another computer.
• Rewrite README.md.
• Save notebooks with images.
• Add github badges.
• Update github and go public.
• Update arxiv link.
• Update CITE.bib and doi badge.
• Make repo for big files and link it in README.md.
• Update article link and doi badge.