Entangled Conditional Adversarial Autoencoder for de Novo Drug Discovery

TLDR

Modern computational approaches and machine learning accelerate new drug invention, and generative models can discover novel molecules within hours versus months for conventional pipelines. The authors propose an entangled conditional adversarial autoencoder that generates molecular structures conditioned on properties such as activity, solubility, or synthetic accessibility. The model was used to generate a novel Janus kinase 3 inhibitor relevant to rheumatoid arthritis, psoriasis, and vitiligo. In vitro testing demonstrated that the generated molecule exhibited good activity and selectivity.

Abstract

Modern computational approaches and machine learning techniques accelerate the invention of new drugs. Generative models can discover novel molecular structures within hours, while conventional drug discovery pipelines require months of work. In this article, we propose a new generative architecture, entangled conditional adversarial autoencoder, that generates molecular structures based on various properties, such as activity against a specific protein, solubility, or ease of synthesis. We apply the proposed model to generate a novel inhibitor of Janus kinase 3, implicated in rheumatoid arthritis, psoriasis, and vitiligo. The discovered molecule was tested in vitro and showed good activity and selectivity.