The evolution, evolvability and engineering of gene regulatory DNA

     

Quickstart

Use the live app now. No downloads. No installation. 👇

Repository overview

The GitHub repository is organized into two directories :

• app : A fully self-contained directory for running the interactive application to compute mutational robustness, evolvability vectors and expression.

• manuscript_code : The codebase corresponding to the manuscript. The organization of this directory is further described here.

Run the app locally

If you wish to run the app on your local machine or cluster,

1. Install Docker.
2. Run the following commands on a terminal :

docker pull edv123456789/evolution_app:latest

docker run --rm -d  -p 8501:8501/tcp edv123456789/evolution_app:latest

The app is now running and you can access it by navigating to http://localhost:8501/ in your web browser. If running on a remote cluster, you may want to expose port 8501 using ngrok.

Using the model directly

1. After installing docker and pulling the latest image as described in the first two steps above, run the following on a terminal :

docker run -it --rm --entrypoint /bin/bash edv123456789/evolution_app

python

2. In the python shell, run :

from app_aux import *

model_condition = 'Glu' #or, 'SC_Ura'

model, _ , __ = load_model(model_condition) 

model.summary()

You have now loaded our tensorflow.keras model. You may use this as is for downstream computations as described in the manuscript or adapt it for your application (e.g. transfer learning).

To exit the python shell and the docker container, simply press Ctrl+D twice.

Data

Download all data and models used in the manuscript here 👇

Interactive Code

The code is also available in an interactive, fully functional form as a CodeOcean capsule (along with all associated data) at :

Reference

The evolution, evolvability and engineering of gene regulatory DNA, Nature 2022.

Eeshit Dhaval Vaishnav^*§, Carl G. de Boer^*§, Jennifer Molinet, Moran Yassour, Lin Fan, Xian Adiconis, Dawn A. Thompson, Joshua Z. Levin, Francisco A. Cubillos, Aviv Regev^§.

DOI : https://doi.org/10.1038/s41586-022-04506-6