An interpretable deep learning framework specifically-designed for cross-species comparison, providing an effective approach for predicting evolutionary cis-regulatory variants.
DeepEvo is implemented in Python and utilizes PyTorch for deep learning. We highly recommend using Conda to manage the environment and ensure reproducibility.
Tested Operating Systems: CentOS Linux 7 (Core)
Typical Install Time: 30-40 minutes (depending on network speed)
git clone https://github.com/bbd0123/DeepEvo.git
cd DeepEvoThis will install Python, PyTorch (v2.4.0), and all required dependencies.
conda env create -f environment.yml
conda activate deepevo_env| Name | Description |
|---|---|
| Demo_code | A step-by-step tutorial demonstrating expression difference prediction and base-level effect inference using ISM with DeepEvo. |
| Demo_data | Example datasets used in the demonstration code. |
| trained_models | Two pretrained models generated using DeepEvo. |
| training_script | Scripts for training DeepEvo models from scratch. |
(Note: The default environment.yml is configured for GPUs with CUDA 11.8. If you are running on a CPU-only machine or a different CUDA version, please refer to the PyTorch official website to modify the pytorch-cuda version accordingly.)