Overcoming DMTA Cycle Challenges: A Unified AI-Driven System for Efficient Drug Design

The integration of artificial intelligence (AI) and machine learning (ML) in drug design has the potential to transform small molecule drug discovery in the pharmaceutical industry, enhancing the efficiency and productivity of the drug design and discovery process. However, the manual and segmented nature of the Design, Make, Test, Analyze (DMTA) cycle is a major obstacle to any significant progress being made. The design of synthetically tractable molecules that meet project specific criteria requires a comprehensive system capable of accounting simultaneously for all synthetic constraints as well as bioactivity and physicochemical properties in order to reach an optimal outcome. The development of such an AI system is complex, requiring the integration of diverse technologies and expertise synergistically. This article outlines our vision and efforts towards the development of a unified system in which AI, ML, computational chemistry, organic chemistry, biology and human expertise converge to drive innovation in drug discovery.