The integration of artificial intelligence (AI) into drug discovery has evolved from early computer-aided design to advanced AI-driven methodologies, laying the foundation for a transformative paradigm: Silico-driven Drug Discovery (SDD). Unlike conventional approaches where AI supports isolated stages, SDD treats the entire research process, including literature understanding, hypothesis generation, molecular design, and experimental validation, as a unified, potentially autonomous system. This review proposes the THINK–BUILD–OPERATE (TBO) architecture as a universal framework for implementing SDD and outlines its six-level automation pathway from human-led to fully autonomous discovery. We highlight nanomedicine as an optimal frontier for SDD due to its well-defined theoretical foundations, abundant multi-omics and pharmacological data, and supportive regulatory shifts. By integrating domain-specific toolchains, large-scale AI models, and orchestrated self-driving laboratories, SDD can accelerate complex, multidisciplinary research while reducing costs and timelines. We further identify key challenges including AI model reliability, infrastructure interoperability, and automated laboratory versatility that must be addressed to achieve this vision. Ultimately, the convergence of AI, advanced laboratory automation, and global research networking holds the potential to transform drug discovery into an industrial-scale, programmable scientific enterprise.
扫码关注我们
求助内容:
应助结果提醒方式:
