Dear User,

This directory contains the code for our manuscript :

The evolution, evolvability, and engineering of gene regulatory DNA

All data and trained models used in the manuscript can be downloaded : 🗄️.

The conda environment required for running this notebook can be installed and activated by running the following on the command line from within this folder:

conda env create -f evolution_env.yml 
conda activate me

This directory is organized into the following subdirectories :

• model : This directory contains all of the code for (training and using) the various models (transformer(tpu), convolutional(gpu), benchmarking models) used in our paper. The README files and Jupyter notebooks in each informatively named subdirectory contain more details.
• ecc_mr_fr: contains the notebooks for computing the Expression Conservation Coefficient (ECC), Mutational Robustness and Fitness Responsivity.
• evolvability : contains the notebooks for computing the evolvability vectors and their two-dimensional representations. It also has examples for generating the landscape visualizations shown in the manuscript that the user may use for their own new sequences.
• ga : contains the notebooks for sequence design using a genetic algorithm.
• rna_seq : contains all the code used for processing the in-house cross-species RNA-seq data.
• trajectories : contains the notebooks for computing the sequence trajectories under different evolutionary regimes.
• regulatory_complexity : points to the repository containing the code for computing the regulatory complexity.