Binding Affinity Prediction Using Self-supervised Learning and QP-Ensemble
2
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
The drug discovery and optimization of candidate compounds are key initial stages in drug development. Predicting the affinity between drugs and proteins, known as drug-target binding affinity (DTA), is a significant problem in the fields of biochemistry, structural biology, and artificial intelligence. In this study, we propose the utilization of ensemble learning with convolutional neural networks (CNNs) using quadratic programming (QP) to improve prediction accuracy. Additionally, we suggest incorporating self-supervised and semi-supervised learning with data augmentation. Experimental results using the Davis dataset demonstrated that the proposed self-supervised learning model outperformed other learning methods in predicting DTA across all four metrics, including mean squared error (MSE).
Abstract
The drug discovery and optimization of candidate compounds are key initial stages in drug development. Predicting the affinity between drugs and proteins, known as drug-target binding affinity (DTA), is a significant problem in the fields of biochemistry, structural biology, and artificial intelligence. In this study, we propose the utilization of ensemble learning with convolutional neural networks (CNNs) using quadratic programming (QP) to improve prediction accuracy. Additionally, we suggest incorporating self-supervised and semi-supervised learning with data augmentation. Experimental results using the Davis dataset demonstrated that the proposed self-supervised learning model outperformed other learning methods in predicting DTA across all four metrics, including mean squared error (MSE).
참고문헌 (9)
- 최연지, "Machine Learning Algorithms for Predicting the Returns of IPO Companies", Quantitative Bio-Science, 40(2), pp.71~75 (2021), 자연과학연구소[DOI]
- 심주용, "Varying Coefficient Regression Modeling using Deep Neural Network", Quantitative Bio-Science, 41(2), pp.71~76 (2022), 자연과학연구소[DOI]
- Pahikkala T, "Toward more realistic drug-target interaction predictions", Brief Bioinform, 16, pp.325~337 (2015)
- Shim J, "Prediction of drug-target binding affinity using similarity-based convolutional neural network", Sci Rep, 11, pp.4416~ (2021)
- Öztürk H, "DeepDTA : deep drug-target binding affinity prediction", Bioinformatics, 34, pp.i821~i829 (2018)[DOI]
- Zhao L, "GANsDTA : Predicting drug-target binding affinity using GANs", Front Genet, 10, pp.1243~ (2020)
- Abbasi K, "DeepCDA : deep cross-domain compound-protein affinity prediction through LSTM and convolutional neural networks", Bioinformatics, 36, pp.4633~4642 (2020)
- He T, "SimBoost : a read-across approach for predicting drug-target binding affinities using gradient boosting machines", J Cheminform, 9, pp.1~14 (2017)
- Pratim P, "On two novel parameters for validation of predictive QSAR models", Molecules, 14, pp.1660~1701 (2009)