ScaffoldGPT: A Scaffold-based GPT Model for Drug Optimization

TL;DR

ScaffoldGPT is a novel transformer-based model designed for drug optimization that effectively balances preserving original scaffolds with enhancing desired drug properties, outperforming existing methods on COVID and cancer benchmarks.

Contribution

The paper introduces ScaffoldGPT, a three-stage optimization approach with a new pre-training strategy and token-level decoding, advancing scaffold-based drug design.

Findings

Outperforms baselines in drug optimization benchmarks

Preserves original functional scaffolds effectively

Enhances desired drug properties in experiments

Abstract

Drug optimization has become increasingly crucial in light of fast-mutating virus strains and drug-resistant cancer cells. Nevertheless, it remains challenging as it necessitates retaining the beneficial properties of the original drug while simultaneously enhancing desired attributes beyond its scope. In this work, we aim to tackle this challenge by introducing ScaffoldGPT, a novel Generative Pretrained Transformer (GPT) designed for drug optimization based on molecular scaffolds. Our work comprises three key components: (1) A three-stage drug optimization approach that integrates pretraining, finetuning, and decoding optimization. (2) A novel two-phase incremental pre-training strategy for scaffold-based drug optimization. (3) A token-level decoding optimization strategy, Top-N, that enabling controlled, reward-guided generation using the pretrained or finetuned GPT. We demonstrate via a comprehensive evaluation on COVID and cancer benchmarks that ScaffoldGPT outperforms the competing baselines in drug optimization benchmarks, while excelling in preserving original functional scaffold and enhancing desired properties.