Hepatotoxicity Modeling Using Counter-Propagation Artificial Neural Networks: Handling an Imbalanced Classification Problem
Drug-induced liver injury is a major concern in the drug development process. Expensive and time-consuming <i>in vitro</i> and <i>in vivo</i> studies do not reflect the complexity of the phenomenon. Complementary to wet lab methods are <i>in silico</i> approaches,...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-01-01
|
| Series: | Molecules |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1420-3049/25/3/481 |
| _version_ | 1818277616974561280 |
|---|---|
| author | Benjamin Bajželj Viktor Drgan |
| author_facet | Benjamin Bajželj Viktor Drgan |
| author_sort | Benjamin Bajželj |
| collection | DOAJ |
| description | Drug-induced liver injury is a major concern in the drug development process. Expensive and time-consuming <i>in vitro</i> and <i>in vivo</i> studies do not reflect the complexity of the phenomenon. Complementary to wet lab methods are <i>in silico</i> approaches, which present a cost-efficient method for toxicity prediction. The aim of our study was to explore the capabilities of counter-propagation artificial neural networks (CPANNs) for the classification of an imbalanced dataset related to idiosyncratic drug-induced liver injury and to develop a model for prediction of the hepatotoxic potential of drugs. Genetic algorithm optimization of CPANN models was used to build models for the classification of drugs into hepatotoxic and non-hepatotoxic class using molecular descriptors. For the classification of an imbalanced dataset, we modified the classical CPANN training algorithm by integrating random subsampling into the training procedure of CPANN to improve the classification ability of CPANN. According to the number of models accepted by internal validation and according to the prediction statistics on the external set, we concluded that using an imbalanced set with balanced subsampling in each learning epoch is a better approach compared to using a fixed balanced set in the case of the counter-propagation artificial neural network learning methodology. |
| first_indexed | 2024-12-12T23:04:23Z |
| format | Article |
| id | doaj.art-b4513ba37f2e423e8906c9e4fbed32ac |
| institution | Directory Open Access Journal |
| issn | 1420-3049 |
| language | English |
| last_indexed | 2024-12-12T23:04:23Z |
| publishDate | 2020-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Molecules |
| spelling | doaj.art-b4513ba37f2e423e8906c9e4fbed32ac2022-12-22T00:08:44ZengMDPI AGMolecules1420-30492020-01-0125348110.3390/molecules25030481molecules25030481Hepatotoxicity Modeling Using Counter-Propagation Artificial Neural Networks: Handling an Imbalanced Classification ProblemBenjamin Bajželj0Viktor Drgan1National Institute of Chemistry, Hajdrihova 19, 1001 Ljubljana, SloveniaNational Institute of Chemistry, Hajdrihova 19, 1001 Ljubljana, SloveniaDrug-induced liver injury is a major concern in the drug development process. Expensive and time-consuming <i>in vitro</i> and <i>in vivo</i> studies do not reflect the complexity of the phenomenon. Complementary to wet lab methods are <i>in silico</i> approaches, which present a cost-efficient method for toxicity prediction. The aim of our study was to explore the capabilities of counter-propagation artificial neural networks (CPANNs) for the classification of an imbalanced dataset related to idiosyncratic drug-induced liver injury and to develop a model for prediction of the hepatotoxic potential of drugs. Genetic algorithm optimization of CPANN models was used to build models for the classification of drugs into hepatotoxic and non-hepatotoxic class using molecular descriptors. For the classification of an imbalanced dataset, we modified the classical CPANN training algorithm by integrating random subsampling into the training procedure of CPANN to improve the classification ability of CPANN. According to the number of models accepted by internal validation and according to the prediction statistics on the external set, we concluded that using an imbalanced set with balanced subsampling in each learning epoch is a better approach compared to using a fixed balanced set in the case of the counter-propagation artificial neural network learning methodology.https://www.mdpi.com/1420-3049/25/3/481hepatotoxicitycounter-propagation artificial neural networksimbalanced datasetgenetic algorithmqsar |
| spellingShingle | Benjamin Bajželj Viktor Drgan Hepatotoxicity Modeling Using Counter-Propagation Artificial Neural Networks: Handling an Imbalanced Classification Problem Molecules hepatotoxicity counter-propagation artificial neural networks imbalanced dataset genetic algorithm qsar |
| title | Hepatotoxicity Modeling Using Counter-Propagation Artificial Neural Networks: Handling an Imbalanced Classification Problem |
| title_full | Hepatotoxicity Modeling Using Counter-Propagation Artificial Neural Networks: Handling an Imbalanced Classification Problem |
| title_fullStr | Hepatotoxicity Modeling Using Counter-Propagation Artificial Neural Networks: Handling an Imbalanced Classification Problem |
| title_full_unstemmed | Hepatotoxicity Modeling Using Counter-Propagation Artificial Neural Networks: Handling an Imbalanced Classification Problem |
| title_short | Hepatotoxicity Modeling Using Counter-Propagation Artificial Neural Networks: Handling an Imbalanced Classification Problem |
| title_sort | hepatotoxicity modeling using counter propagation artificial neural networks handling an imbalanced classification problem |
| topic | hepatotoxicity counter-propagation artificial neural networks imbalanced dataset genetic algorithm qsar |
| url | https://www.mdpi.com/1420-3049/25/3/481 |
| work_keys_str_mv | AT benjaminbajzelj hepatotoxicitymodelingusingcounterpropagationartificialneuralnetworkshandlinganimbalancedclassificationproblem AT viktordrgan hepatotoxicitymodelingusingcounterpropagationartificialneuralnetworkshandlinganimbalancedclassificationproblem |