RESUMO
Recently emerging generative AI models enable us to produce a vast number of compounds for potential applications. While they can provide novel molecular structures, the synthetic feasibility of the generated molecules is often questioned. To address this issue, a few recent studies have attempted to use deep learning models to estimate the synthetic accessibility of many molecules rapidly. However, retrosynthetic analysis tools used to train the models rely on reaction templates automatically extracted from a large reaction database that are not domain-specific and may exhibit low chemical correctness. To overcome this limitation, we introduce DFRscore (Drug-Focused Retrosynthetic score), a deep learning-based approach for a more practical assessment of synthetic accessibility in drug discovery. The DFRscore model is trained exclusively on drug-focused reactions, providing a predicted number of minimally required synthetic steps for each compound. This approach enables practitioners to filter out compounds that do not meet their desired level of synthetic accessibility at an early stage of high-throughput virtual screening for accelerated drug discovery. The proposed strategy can be easily adapted to other domains by adjusting the synthesis planning setup of the reaction templates and starting materials.
Assuntos
Aprendizado Profundo , Efeitos Colaterais e Reações Adversas Relacionados a Medicamentos , Humanos , Descoberta de Drogas , Ensaios de Triagem em Larga Escala , Estrutura Molecular , Bases de Dados FactuaisRESUMO
Securinega alkaloids have fascinated the synthetic chemical community for over six decades. Historically, major research foci in securinega alkaloid synthesis have been on the efficient construction of the fused tetracyclic framework that bears a butenolide moiety and tertiary amine-based heterocycles. These "basic" securinega alkaloids have evolved to undergo biosynthetic oxidative diversifications, especially on the piperidine core. However, a general synthetic solution to access these high-oxidation state securinega alkaloids is lacking. In this study, we have completed the total synthesis of various C4-oxygenated securinine-type alkaloids including securingines A, C, D, securitinine, secu'amamine D, phyllanthine, and 4-epi-phyllanthine. Our synthetic strategy features stereocontrolled oxidation, rearrangement, and epimerization at N1 and C2-C4 positions of the piperidine core within (neo)securinane scaffolds. Our discoveries provide a fundamental synthetic solution to all known securinine-type natural products with various oxidative and stereochemical variations around the central piperidine ring.