Drug Molecule Generation (DMG)

This project implements a Variational Autoencoder (VAE) for generating drug molecules. The VAE model is designed to encode molecular structures into a latent space and decode them back into molecular structures. Additionally, the model predicts properties of the generated molecules.

Setup

Environment

First create new virtual environment.

python -m venv .venv
source .venv/bin/activate
pip install .

Usage

dmg --help

Relevant Links

https://keras.io/examples/generative/molecule_generation/

https://raw.githubusercontent.com/aspuru-guzik-group/chemical_vae/master/models/zinc_properties/250k_rndm_zinc_drugs_clean_3.csv

https://huggingface.co/keras-io/drug-molecule-generation-with-VAE

Relevant Papers

Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules

MolGAN: An implicit generative model for small molecular graphs