Methods Mol Biol. 2026;3061:25-38. doi: 10.1007/978-1-0716-5420-0_2.
ABSTRACT
Drug design is one of the most resource-intensive and failure-prone stages of pharmaceutical R&D. This chapter presents the perspective of computational chemists at Sygnature Discovery on the growing need to integrate quantum mechanics (QM) with artificial intelligence (AI) to improve the efficiency and effectiveness of modern drug design. Published retrospective analyses and case studies indicate that programs supported by computational chemistry can progress more efficiently through the design phase while reducing experimental burden, including synthetic effort. Structure-based drug design (SBDD) remains the dominant computational approach; however, its accuracy is constrained by simplified scoring functions. QM, particularly the Fragment Molecular Orbital (FMO) method, addresses some of these limitations by providing residue-level, experimentally aligned energetic insight. AI methods have expanded rapidly but remain constrained by data quality, interpretability, and the practical challenge of experimentally validating large numbers of generated designs. In this chapter, we discuss the perspectives and opportunities of combining AI and QM within SBDD to create more predictive, efficient, and mechanistically informed drug design workflows.
PMID:42604895 | DOI:10.1007/978-1-0716-5420-0_2