Molecular generative model based on conditional variational autoencoder for de novo molecular design

TL;DR

This paper introduces a conditional variational autoencoder-based molecular generative model capable of designing drug-like molecules with multiple controllable properties, demonstrating flexibility in property manipulation beyond dataset ranges.

Contribution

It presents a novel conditional VAE model that controls multiple molecular properties simultaneously for de novo molecular design.

Findings

Successfully generated drug-like molecules with five target properties.

Able to adjust individual properties without affecting others.

Manipulated properties beyond dataset ranges.

Abstract

We propose a molecular generative model based on the conditional variational autoencoder for de novo molecular design. It is specialized to control multiple molecular properties simultaneously by imposing them on a latent space. As a proof of concept, we demonstrate that it can be used to generate drug-like molecules with five target properties. We were also able to adjust a single property without changing the others and to manipulate it beyond the range of the dataset.