ISSN 1000-1239 CN 11-1777/TP

计算机研究与发展 ›› 2022, Vol. 59 ›› Issue (9): 2051-2065.doi: 10.7544/issn1000-1239.20210134

• 人工智能 • 上一篇    下一篇

注意力特征融合的蛋白质-药物相互作用预测

华阳1,李金星1,冯振华2,宋晓宁1,孙俊1,於东军3   

  1. 1(江南大学人工智能与计算机学院 江苏无锡 214100);2(萨里大学计算机系 英国吉尔福德 GU2 7XH);3(南京理工大学计算机科学与工程学院 南京 210094) (7211905018@stu.jiangnan.edu.cn)
  • 出版日期: 2022-09-01
  • 基金资助: 
    国家重点研发计划项目(2017YFC1601800);国家自然科学基金项目(61876072,61902153,62072243,61772273);中国博士后科学基金项目(2018T110441);江苏省六大人才高峰项目(XYDXX-012)

Protein-Drug Interaction Prediction Based on Attention Feature Fusion

Hua Yang1, Li Jinxing1, Feng Zhenhua2, Song Xiaoning1, Sun Jun1, Yu Dongjun3   

  1. 1(School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, Jiangsu 214100);2(Department of Computer Science, University of Surrey, Guildford, UK GU2 7XH);3(School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094)
  • Online: 2022-09-01
  • Supported by: 
    This work was supported by the National Key Research and Development Program(2017YFC1601800), the National Natural Science Foundation of China(61876072, 61902153,62072243,61772273), the China Postdoctoral Science Foundation(2018T110441), and the Six Talent Peaks Project of Jiangsu Province (XYDXX-012).

摘要: 药物一般通过抑制或激活人体中某些蛋白活性反应进而发挥效能,因此预测蛋白和药物的相互作用对新药开发的筛选工作极为关键.然而,基于传统方法在湿实验中进行该类实验需要耗费巨大的人力和物力.为解决这一问题,提出了一种基于自注意力机制和多药物特征融合的蛋白质-药物相互作用预测算法.首先,合理融合基于药物分子结构特征的Morgan指纹、Mol2Vec表示向量以及消息传递网络所提特征;随后,将融合结果对由密集型卷积所提取的蛋白特征做注意力加权;接着综合两者特征,利用自注意力机制和双向门控循环单元预测蛋白质药物相互作用;最后,根据训练模型设计了可应用的预测系统,并展示了其在筛选治愈阿尔兹海默症药物的具体使用方法和效果.实验结果表明,较现有的预测方法,新方法在BindingDB,Kinase,Human,C.elegans数据集上均达到了更好的预测效果.最优的AUC分别达到了0.963,0.937,0.983,0.990,较同类方法具有十分明显的优势.

关键词: 蛋白质-药物相互作用, 药物筛选, 特征融合, 密集型卷积网络, 自注意力机制

Abstract: Drugs usually work by inhibiting or activating the active reactions of certain proteins in the human body, so the prediction of the interactions between proteins and drugs is very important for the screening of new drugs. However, it takes a lot of manpower and material resources to carry out this kind of wet experiment using traditional methods. To resolve this problem, we propose a protein-drug interaction prediction algorithm based on the self-attention mechanism and multi-drug feature fusion. Firstly, the Morgan fingerprint based on drug molecular structure characteristics, the Mol2Vec representation vector, and the features extracted by the messaging network are reasonably fused. Secondly, the fusion results are used to weigh the protein features extracted by dense convolution. After that, the self-attentional mechanism and bidirectional gating circulatory unit are used to predict protein-drug interactions by combining their characteristics. Finally, an applicable prediction system based on the training model is designed, which demonstrates the specific use cases and effects of the proposed method in drug screening for the Alzheimer disease. The experimental results show that the proposed algorithm achieves better prediction performance on BindingDB, Kinase, Human and C.elegans datasets compared with the existing prediction methods. The AUC values achieve 0.963, 0.937, 0.983, 0.990 on the four datasets, demonstrating significant superiority over the other algorithms.

Key words: protein-drug interaction, drug screening, feature fusion, DenseNet, self-attention mechanism

中图分类号: