An Effective Pipeline for Training Variational Autoencoders for Synthesizable and Optimized Molecular Design
TL;DR
This article presents a pipeline for training variational autoencoders to design synthesizable and optimized molecules, enhancing drug discovery by improving properties like affinity and bioavailability.
An Effective Pipeline for Training Variational Autoencoders for Synthesizable and Optimized Molecular Design
Fardeen H. Mozumder; Byung-Jun Yoon
https://doi.org/10.1109/ACCESS.2024.3523531
Volume 13
Variational auto-encoders (VAE) for molecular design and optimization have gained popularity due to their efficiency in exploring high-dimensional molecular space to identify novel molecules with various properties of interest. For example, when applied to drug discovery, one may want to optimize small molecules for higher affinity against a specific target, improved bioavailability, and solubilit...