UnbiasedDTI: Mitigating Real-World Bias of Drug-Target Interaction Prediction by Using Deep Ensemble-Balanced Learning

被引:9
作者
Tayebi, Aida [1 ]
Yousefi, Niloofar [1 ]
Yazdani-Jahromi, Mehdi [1 ]
Kolanthai, Elayaraja [2 ]
Neal, Craig J. [2 ]
Seal, Sudipta [2 ,3 ]
Garibay, Ozlem Ozmen [1 ]
机构
[1] Univ Cent Florida, Dept Ind Engn & Management Syst, Orlando, FL 32816 USA
[2] Univ Cent Florida, Adv Mat Proc & Anal Ctr, Dept Mat Sci & Engn, Orlando, FL 32816 USA
[3] Univ Cent Florida, Coll Med, Bionix Cluster, Orlando, FL 32816 USA
来源
MOLECULES | 2022年 / 27卷 / 09期
关键词
drug-target interaction; ensemble learning; deep learning; machine learning; spike protein; SARS-CoV-2; ACE2; receptor; IDENTIFICATION; EVOLUTIONARY;
D O I
10.3390/molecules27092980
中图分类号
Q5 [生物化学]; Q7 [分子生物学];
学科分类号
071010 ; 081704 ;
摘要
Drug-target interaction (DTI) prediction through in vitro methods is expensive and time-consuming. On the other hand, computational methods can save time and money while enhancing drug discovery efficiency. Most of the computational methods frame DTI prediction as a binary classification task. One important challenge is that the number of negative interactions in all DTI-related datasets is far greater than the number of positive interactions, leading to the class imbalance problem. As a result, a classifier is trained biased towards the majority class (negative class), whereas the minority class (interacting pairs) is of interest. This class imbalance problem is not widely taken into account in DTI prediction studies, and the few previous studies considering balancing in DTI do not focus on the imbalance issue itself. Additionally, they do not benefit from deep learning models and experimental validation. In this study, we propose a computational framework along with experimental validations to predict drug-target interaction using an ensemble of deep learning models to address the class imbalance problem in the DTI domain. The objective of this paper is to mitigate the bias in the prediction of DTI by focusing on the impact of balancing and maintaining other involved parameters at a constant value. Our analysis shows that the proposed model outperforms unbalanced models with the same architecture trained on the BindingDB both computationally and experimentally. These findings demonstrate the significance of balancing, which reduces the bias towards the negative class and leads to better performance. It is important to note that leaning on computational results without experimentally validating them and by relying solely on AUROC and AUPRC metrics is not credible, particularly when the testing set remains unbalanced.
引用
收藏
页数:16
相关论文
共 50 条
  • [41] A deep learning method for drug-target affinity prediction based on sequence interaction information mining
    Jiang, Mingjian
    Shao, Yunchang
    Zhang, Yuanyuan
    Zhou, Wei
    Pang, Shunpeng
    PEERJ, 2023, 11
  • [42] A deep learning method for drug-target affinity prediction based on sequence interaction information mining
    Jiang, Mingjian
    Shao, Yunchang
    Zhang, Yuanyuan
    Zhou, Wei
    Pang, Shunpeng
    PEERJ, 2023, 11
  • [43] GSRF-DTI: a framework for drug-target interaction prediction based on a drug-target pair network and representation learning on a large graph
    Zhu, Yongdi
    Ning, Chunhui
    Zhang, Naiqian
    Wang, Mingyi
    Zhang, Yusen
    BMC BIOLOGY, 2024, 22 (01)
  • [44] A Machine Learning Approach for Drug-Target Interaction Prediction using Wrapper Feature Selection and Class Balancing
    Redkar, Shweta
    Mondal, Sukanta
    Joseph, Alex
    Hareesha, K. S.
    MOLECULAR INFORMATICS, 2020, 39 (05)
  • [45] Drug-Target Interaction Prediction Model Using Optimal Recurrent Neural Network
    Kavipriya, G.
    Manjula, D.
    INTELLIGENT AUTOMATION AND SOFT COMPUTING, 2023, 35 (02) : 1675 - 1689
  • [46] Drug-Online: an online platform for drug-target interaction, affinity, and binding sites identification using deep learning
    Zeng, Xin
    Su, Guang-Peng
    Li, Shu-Juan
    Lv, Shuang-Qing
    Wen, Meng-Liang
    Li, Yi
    BMC BIOINFORMATICS, 2024, 25 (01)
  • [47] Prediction of drug-target binding affinity based on deep learning models
    Zhang H.
    Liu X.
    Cheng W.
    Wang T.
    Chen Y.
    Computers in Biology and Medicine, 2024, 174
  • [48] QLDTI: A Novel Reinforcement Learning-based Prediction Model for Drug-Target Interaction
    Gao, Jie
    Fu, Qiming
    Sun, Jiacheng
    Wang, Yunzhe
    Xia, Youbing
    Lu, You
    Wu, Hongjie
    Chen, Jianping
    CURRENT BIOINFORMATICS, 2024, 19 (04) : 352 - 374
  • [49] Prediction of Drug-Target Interactions by Ensemble Learning Method From Protein Sequence and Drug Fingerprint
    Zhan, Xinke
    You, Zhu-Hong
    Cai, Jinfan
    Li, Liping
    Yu, Changqing
    Pan, Jie
    Kong, Jiangkun
    IEEE ACCESS, 2020, 8 : 185465 - 185476
  • [50] DeepFusionDTA: Drug-Target Binding Affinity Prediction With Information Fusion and Hybrid Deep-Learning Ensemble Model
    Pu, Yuqian
    Li, Jiawei
    Tang, Jijun
    Guo, Fei
    IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS, 2022, 19 (05) : 2760 - 2769