Molecular Variational Autoencoder
Deep Learning for Drug Discovery
Overview
Implemented a Tensorflow version of the Molecular VAE paper for automated molecular design. Trained on a dataset of 250,000 molecular graphs from the ZINC database, this deep learning variational autoencoder learns to encode molecular structures into a continuous latent space, enabling the generation of novel drug-like molecules. The model facilitates faster drug development by allowing researchers to explore chemical space more efficiently and generate molecules with desired properties.
Key Features
Variational Autoencoder
Encodes molecular structures into continuous latent representations for generation
Large Scale Training
Trained on 250,000 molecular graphs from the ZINC database
Molecule Generation
Generates novel drug-like molecules by sampling from latent space
Tech Stack
Project Info
Contribution
Independent project spanning product decisions and implementation.