A Novel Deep Neural Network Technique for Drug–Target Interaction
Drug discovery (DD) is a time-consuming and expensive process. Thus, the industry employs strategies such as drug repositioning and drug repurposing, which allows the application of already approved drugs to treat a different disease, as occurred in the first months of 2020, during the COVID-19 pand...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-03-01
|
| Series: | Pharmaceutics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4923/14/3/625 |
| _version_ | 1797443537519771648 |
|---|---|
| author | Jackson G. de Souza Marcelo A. C. Fernandes Raquel de Melo Barbosa |
| author_facet | Jackson G. de Souza Marcelo A. C. Fernandes Raquel de Melo Barbosa |
| author_sort | Jackson G. de Souza |
| collection | DOAJ |
| description | Drug discovery (DD) is a time-consuming and expensive process. Thus, the industry employs strategies such as drug repositioning and drug repurposing, which allows the application of already approved drugs to treat a different disease, as occurred in the first months of 2020, during the COVID-19 pandemic. The prediction of drug–target interactions is an essential part of the DD process because it can accelerate it and reduce the required costs. DTI prediction performed <i>in silico</i> have used approaches based on molecular docking simulations, including similarity-based and network- and graph-based ones. This paper presents MPS2IT-DTI, a DTI prediction model obtained from research conducted in the following steps: the definition of a new method for encoding molecule and protein sequences onto images; the definition of a deep-learning approach based on a convolutional neural network in order to create a new method for DTI prediction. Training results conducted with the Davis and KIBA datasets show that MPS2IT-DTI is viable compared to other state-of-the-art (SOTA) approaches in terms of performance and complexity of the neural network model. With the Davis dataset, we obtained 0.876 for the concordance index and 0.276 for the MSE; with the KIBA dataset, we obtained 0.836 and 0.226 for the concordance index and the MSE, respectively. Moreover, the MPS2IT-DTI model represents molecule and protein sequences as images, instead of treating them as an NLP task, and as such, does not employ an embedding layer, which is present in other models. |
| first_indexed | 2024-03-09T12:57:32Z |
| format | Article |
| id | doaj.art-1140875945334d8889c5c3b82681513a |
| institution | Directory Open Access Journal |
| issn | 1999-4923 |
| language | English |
| last_indexed | 2024-03-09T12:57:32Z |
| publishDate | 2022-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Pharmaceutics |
| spelling | doaj.art-1140875945334d8889c5c3b82681513a2023-11-30T21:57:35ZengMDPI AGPharmaceutics1999-49232022-03-0114362510.3390/pharmaceutics14030625A Novel Deep Neural Network Technique for Drug–Target InteractionJackson G. de Souza0Marcelo A. C. Fernandes1Raquel de Melo Barbosa2Laboratory of Machine Learning and Intelligent Instrumentation, Federal University of Rio Grande do Norte, Natal 59078-970, BrazilLaboratory of Machine Learning and Intelligent Instrumentation, Federal University of Rio Grande do Norte, Natal 59078-970, BrazilLaboratory of Machine Learning and Intelligent Instrumentation, Federal University of Rio Grande do Norte, Natal 59078-970, BrazilDrug discovery (DD) is a time-consuming and expensive process. Thus, the industry employs strategies such as drug repositioning and drug repurposing, which allows the application of already approved drugs to treat a different disease, as occurred in the first months of 2020, during the COVID-19 pandemic. The prediction of drug–target interactions is an essential part of the DD process because it can accelerate it and reduce the required costs. DTI prediction performed <i>in silico</i> have used approaches based on molecular docking simulations, including similarity-based and network- and graph-based ones. This paper presents MPS2IT-DTI, a DTI prediction model obtained from research conducted in the following steps: the definition of a new method for encoding molecule and protein sequences onto images; the definition of a deep-learning approach based on a convolutional neural network in order to create a new method for DTI prediction. Training results conducted with the Davis and KIBA datasets show that MPS2IT-DTI is viable compared to other state-of-the-art (SOTA) approaches in terms of performance and complexity of the neural network model. With the Davis dataset, we obtained 0.876 for the concordance index and 0.276 for the MSE; with the KIBA dataset, we obtained 0.836 and 0.226 for the concordance index and the MSE, respectively. Moreover, the MPS2IT-DTI model represents molecule and protein sequences as images, instead of treating them as an NLP task, and as such, does not employ an embedding layer, which is present in other models.https://www.mdpi.com/1999-4923/14/3/625drug–target interactionDTI predictiondeep learningconvolutional neural network |
| spellingShingle | Jackson G. de Souza Marcelo A. C. Fernandes Raquel de Melo Barbosa A Novel Deep Neural Network Technique for Drug–Target Interaction Pharmaceutics drug–target interaction DTI prediction deep learning convolutional neural network |
| title | A Novel Deep Neural Network Technique for Drug–Target Interaction |
| title_full | A Novel Deep Neural Network Technique for Drug–Target Interaction |
| title_fullStr | A Novel Deep Neural Network Technique for Drug–Target Interaction |
| title_full_unstemmed | A Novel Deep Neural Network Technique for Drug–Target Interaction |
| title_short | A Novel Deep Neural Network Technique for Drug–Target Interaction |
| title_sort | novel deep neural network technique for drug target interaction |
| topic | drug–target interaction DTI prediction deep learning convolutional neural network |
| url | https://www.mdpi.com/1999-4923/14/3/625 |
| work_keys_str_mv | AT jacksongdesouza anoveldeepneuralnetworktechniquefordrugtargetinteraction AT marceloacfernandes anoveldeepneuralnetworktechniquefordrugtargetinteraction AT raqueldemelobarbosa anoveldeepneuralnetworktechniquefordrugtargetinteraction AT jacksongdesouza noveldeepneuralnetworktechniquefordrugtargetinteraction AT marceloacfernandes noveldeepneuralnetworktechniquefordrugtargetinteraction AT raqueldemelobarbosa noveldeepneuralnetworktechniquefordrugtargetinteraction |