Deep contrastive learning enables genome-wide virtual screening

• 成果类型: Article
• 署名作者: Jia, Yinjun; Gao, Bowen; Tan, Jiaxin; Zheng, Jiqing; Hong, Xin; Zhu, Wenyu; Tan, Haichuan; Xiao, Yuan; Tan, Liping; Cai, Hongyi; Huang, Yanwen; Deng, Zhiheng; Wu, Xiangwei; Jin, Yue; Yuan, Yafei; Tian, Jiekang; He, Wei; Ma, Weiying; Zhang, Yaqin; Liu, Lei; Yan, Chuangye; Zhang, Wei; Lan, Yanyan
• 署名单位: Tsinghua University; Tsinghua University; Tsinghua University; Tsinghua University; Tsinghua University; Tsinghua University; Tsinghua University; Tsinghua University; Tsinghua University; Peking University; Tsinghua University; Beijing Academy of Artificial Intelligence
• 刊物名称: SCIENCE
• ISSN/ISSBN: 0036-8075; 1095-9203
• DOI: 10.1126/science.ads9530
• 发表日期: 2026-01-08
• 页码: eads9530
• 关键词: SCORING FUNCTIONS; neural-network; cryo-em; protein; docking; receptor; database; accuracy; inhibitors; prediction

摘要

Recent breakthroughs in protein structure prediction have opened new avenues for genome-wide drug discovery, yet existing virtual screening methods remain computationally prohibitive. We present DrugCLIP, a contrastive learning framework that achieves ultrafast and accurate virtual screening, up to 10 million times faster than docking, while consistently outperforming various baselines on in silico benchmarks. In wet-lab validations, DrugCLIP achieved a 15% hit rate for norepinephrine transporter, and structures of two identified inhibitors were determined in complex with the target protein. For thyroid hormone receptor interactor 12, a target that lacks holo structures and small-molecule binders, DrugCLIP achieved a 17.5% hit rate using only AlphaFold2-predicted structures. Finally, we released GenomeScreenDB, an open-access database providing precomputed results for similar to 10,000 human proteins screened against 500 million compounds, pioneering a drug discovery paradigm in the post-AlphaFold era.