Drug Target Interaction Prediction using Multi-task Learning and Co-attention

Various machine learning models have been proposed as cost-effective means to predict Drug-Target Interactions (DTI). Most existing researches treat DTI prediction either as a classification task (i.e. output negative or positive labels to indicate existence of interaction) or as a regression task (i.e. output numerical values as the strength of interaction). However, classifiers are more prone to higher bias and regression models tend to overfit the training data to generate large variance. In this paper, we explore to balance the bias and variance by a multi-task learning framework. We propose an architecture to both predict accurate values of strength of interaction and decide correct boundary between positive and negative interactions. Furthermore, the two tasks are performed on a shared feature representation, which is learnt using a co-attention mechanism. Comprehensive experiments demonstrate that the proposed method significantly outperforms state-of-the-art methods.

[1]  Houfeng Wang,et al.  Interactive Attention Networks for Aspect-Level Sentiment Classification , 2017, IJCAI.

[2]  Guillermo Palma,et al.  Drug-Target Interaction Prediction Using Semantic Similarity and Edge Partitioning , 2014, SEMWEB.

[3]  Daniel R. Caffrey,et al.  Structure-based maximal affinity model predicts small-molecule druggability , 2007, Nature Biotechnology.

[4]  Jian Peng,et al.  A Network Integration Approach for Drug-Target Interaction Prediction and Computational Drug Repositioning from Heterogeneous Information , 2017, RECOMB 2017.

[5]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[6]  Yongdong Zhang,et al.  Drug-target interaction prediction: databases, web servers and computational models , 2016, Briefings Bioinform..

[7]  Ming Wen,et al.  Deep-Learning-Based Drug-Target Interaction Prediction. , 2017, Journal of proteome research.

[8]  Jian Peng,et al.  A network integration approach for drug-target interaction prediction and computational drug repositioning from heterogeneous information , 2017, Nature Communications.

[9]  Artem Cherkasov,et al.  SimBoost: a read-across approach for predicting drug–target binding affinities using gradient boosting machines , 2017, Journal of Cheminformatics.

[10]  Heng Luo,et al.  Molecular Docking for Identification of Potential Targets for Drug Repurposing. , 2016, Current topics in medicinal chemistry.

[11]  Dongyan Zhao,et al.  Multi-grained Attention Network for Aspect-Level Sentiment Classification , 2018, EMNLP.

[12]  Arzucan Özgür,et al.  DeepDTA: deep drug–target binding affinity prediction , 2018, Bioinform..

[13]  Yuhao Wang,et al.  Predicting drug-target interactions using restricted Boltzmann machines , 2013, Bioinform..

[14]  Hyeon-Eui Kim,et al.  Deep mining heterogeneous networks of biomedical linked data to predict novel drug‐target associations , 2017, Bioinform..

[15]  Tapio Pahikkala,et al.  Toward more realistic drug^target interaction predictions , 2014 .

[16]  Kwong-Sak Leung,et al.  Low-Quality Structural and Interaction Data Improves Binding Affinity Prediction via Random Forest , 2015, Molecules.

[17]  A. Barabasi,et al.  Network-based in silico drug efficacy screening , 2016, Nature Communications.

[18]  Yonghua Wang,et al.  Pred-binding: large-scale protein–ligand binding affinity prediction , 2016, Journal of enzyme inhibition and medicinal chemistry.

[19]  Michael J. Keiser,et al.  Relating protein pharmacology by ligand chemistry , 2007, Nature Biotechnology.

[20]  Jun Sese,et al.  Compound‐protein interaction prediction with end‐to‐end learning of neural networks for graphs and sequences , 2018, Bioinform..

[21]  Chen Lin,et al.  Drug Target Interaction Prediction with Non-random Missing Labels , 2018, 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[22]  Shujian Huang,et al.  Deep Matrix Factorization Models for Recommender Systems , 2017, IJCAI.

[23]  Bo Liao,et al.  Screening drug-target interactions with positive-unlabeled learning , 2017, Scientific Reports.

[24]  Weiping Chen,et al.  A protein network descriptor server and its use in studying protein, disease, metabolic and drug targeted networks , 2016, Briefings Bioinform..

[25]  Hao Ding,et al.  Similarity-based machine learning methods for predicting drug-target interactions: a brief review , 2014, Briefings Bioinform..