Generative Recurrent Networks for <i>De Novo</i> Drug Design

TLDR

Generative AI models provide a new approach to chemogenomics and de novo drug design by narrowing chemical space to regions of interest. The study introduces a generative RNN–LSTM method for molecular de novo design. The approach trains an RNN–LSTM on SMILES syntax, uses learned pattern probabilities for de novo SMILES generation, and fine‑tunes predictions for specific targets via transfer learning. The model accurately learns SMILES syntax, eliminates library enumeration and external activity prediction, and is advocated for low‑data discovery, fragment design, and hit‑to‑lead optimization.

Abstract

Abstract Generative artificial intelligence models present a fresh approach to chemogenomics and de novo drug design, as they provide researchers with the ability to narrow down their search of the chemical space and focus on regions of interest. We present a method for molecular de novo design that utilizes generative recurrent neural networks (RNN) containing long short‐term memory (LSTM) cells. This computational model captured the syntax of molecular representation in terms of SMILES strings with close to perfect accuracy. The learned pattern probabilities can be used for de novo SMILES generation. This molecular design concept eliminates the need for virtual compound library enumeration. By employing transfer learning, we fine‐tuned the RNN′s predictions for specific molecular targets. This approach enables virtual compound design without requiring secondary or external activity prediction, which could introduce error or unwanted bias. The results obtained advocate this generative RNN‐LSTM system for high‐impact use cases, such as low‐data drug discovery, fragment based molecular design, and hit‐to‐lead optimization for diverse drug targets.