Open-Source Molecular Processing Pipeline for Generating Molecules

TL;DR

This paper introduces an open-source, user-friendly pipeline integrated into DeepChem for molecular generation, featuring high-quality PyTorch implementations of MolGAN and Normalizing Flows that perform comparably to existing methods.

Contribution

It provides a reusable, accessible infrastructure for molecular generative models within DeepChem, including new PyTorch implementations of MolGAN and Normalizing Flows.

Findings

Implementations show strong performance comparable with past work.

Integration into DeepChem enhances accessibility for non-experts.

Open-source pipeline facilitates future research and development.

Abstract

Generative models for molecules have shown considerable promise for use in computational chemistry, but remain difficult to use for non-experts. For this reason, we introduce open-source infrastructure for easily building generative molecular models into the widely used DeepChem [Ramsundar et al., 2019] library with the aim of creating a robust and reusable molecular generation pipeline. In particular, we add high quality PyTorch [Paszke et al., 2019] implementations of the Molecular Generative Adversarial Networks (MolGAN) [Cao and Kipf, 2022] and Normalizing Flows [Papamakarios et al., 2021]. Our implementations show strong performance comparable with past work [Kuznetsov and Polykovskiy, 2021, Cao and Kipf, 2022].