药物反应
药品
抗癌药
计算机科学
人工智能
机器学习
计算生物学
药理学
医学
生物
作者
Xiaoqi Wang,Yiping Wen,Yixin Zhang,Chong Dai,Yaning Yang,Xiaochen Bo,Song He,Shaoliang Peng
摘要
Anticancer drug response prediction with deep learning technology has become the foundation of precision medicine. However, existing methods fail to model both inter- and external relationship of biochemical entities. Therefore, we propose MultiDRP that is a multi-view deep attention network integrating biochemical relationship for anticancer drug response prediction. In external view, we developed cell line and drug network-based graph attention models to capture the association among biological entities. In internal view, self-attention model is used to learn the relationship among feature items in cell lines and drugs. MultiDRP can accurately identify the sensitive and resistant drugs for cancer cell lines. The in vitro experiments for lung cell line NCI-H23 showed that seven out of eight novel drugs predicted by MultiDRP exhibit an IC50 value of less than 10nM. In addition, MultiDRP achieved better performance than six baselines in various prediction scenarios. These results suggest that MultiDRP is a powerful tool for drug response prediction.Funding: This work was supported by NSFC Grants U19A2067 and 62103436; National Key R&D Program of China 2022YFC3400400; The Funds of State Key Laboratory of Chemo/Biosensing and Chemometrics, the National Supercomputing Center in Changsha, and Peng Cheng Lab.Declaration of Interest: All authors declare that they have no competing inter
科研通智能强力驱动
Strongly Powered by AbleSci AI