Abstract
The drug discovery and development process is lengthy and expensive, and bringing a drug to market may take up to 18 years and may cost up to 2 billion $US. The extensive use of computer-assisted drug design techniques may considerably increase the chances of finding valuable drug candidates, thus decreasing the drug discovery time and costs. The most important computational approach is represented by structure-activity relationships that can discriminate between sets of chemicals that are active/inactive towards a certain biological receptor. An adverse effect of some cationic amphiphilic drugs is phospholipidosis that manifests as an intracellular accumulation of phospholipids and formation of concentric lamellar bodies. Here we present structure-activity relationships (SAR) computed with a wide variety of machine learning algorithms trained to identify drugs that have phospholipidosis inducing potential. All SAR models are developed with the machine learning software Weka, and include both classical algorithms, such as k-nearest neighbors and decision trees, as well as recently introduced methods, such as support vector machines and artificial immune systems. The best predictions are obtained with support vector machines, followed by perceptron artificial neural network, logistic regression, and k-nearest neighbors.
Keywords: Machine learning, support vector machines, artificial immune systems, structure-activity relationships, computerassisted drug design, phospholipidosis
Current Topics in Medicinal Chemistry
Title: Weka Machine Learning for Predicting the Phospholipidosis Inducing Potential
Volume: 8 Issue: 18
Author(s): Ovidiu Ivanciuc
Affiliation:
Keywords: Machine learning, support vector machines, artificial immune systems, structure-activity relationships, computerassisted drug design, phospholipidosis
Abstract: The drug discovery and development process is lengthy and expensive, and bringing a drug to market may take up to 18 years and may cost up to 2 billion $US. The extensive use of computer-assisted drug design techniques may considerably increase the chances of finding valuable drug candidates, thus decreasing the drug discovery time and costs. The most important computational approach is represented by structure-activity relationships that can discriminate between sets of chemicals that are active/inactive towards a certain biological receptor. An adverse effect of some cationic amphiphilic drugs is phospholipidosis that manifests as an intracellular accumulation of phospholipids and formation of concentric lamellar bodies. Here we present structure-activity relationships (SAR) computed with a wide variety of machine learning algorithms trained to identify drugs that have phospholipidosis inducing potential. All SAR models are developed with the machine learning software Weka, and include both classical algorithms, such as k-nearest neighbors and decision trees, as well as recently introduced methods, such as support vector machines and artificial immune systems. The best predictions are obtained with support vector machines, followed by perceptron artificial neural network, logistic regression, and k-nearest neighbors.
Export Options
About this article
Cite this article as:
Ivanciuc Ovidiu, Weka Machine Learning for Predicting the Phospholipidosis Inducing Potential, Current Topics in Medicinal Chemistry 2008; 8 (18) . https://dx.doi.org/10.2174/156802608786786589
DOI https://dx.doi.org/10.2174/156802608786786589 |
Print ISSN 1568-0266 |
Publisher Name Bentham Science Publisher |
Online ISSN 1873-4294 |
Call for Papers in Thematic Issues
Chemistry Based on Natural Products for Therapeutic Purposes
The development of new pharmaceuticals for a wide range of medical conditions has long relied on the identification of promising natural products (NPs). There are over sixty percent of cancer, infectious illness, and CNS disease medications that include an NP pharmacophore, according to the Food and Drug Administration. Since NP ...read more
Current Trends in Drug Discovery Based on Artificial Intelligence and Computer-Aided Drug Design
Drug development discovery has faced several challenges over the years. In fact, the evolution of classical approaches to modern methods using computational methods, or Computer-Aided Drug Design (CADD), has shown promising and essential results in any drug discovery campaign. Among these methods, molecular docking is one of the most notable ...read more
Drug Discovery in the Age of Artificial Intelligence
In the age of artificial intelligence (AI), we have witnessed a significant boom in AI techniques for drug discovery. AI techniques are increasingly integrated and accelerating the drug discovery process. These developments have not only attracted the attention of academia and industry but also raised important questions regarding the selection ...read more
From Biodiversity to Chemical Diversity: Focus of Flavonoids
Flavonoids are the largest group of polyphenols, plant secondary metabolites arising from the essential aromatic amino acid phenylalanine (or more rarely from tyrosine) via the phenylpropanoid pathway. The flavan nucleus is the basic 15-carbon skeleton of flavonoids (C6-C3-C6), which consists of two phenyl rings (A and B) and a heterocyclic ...read more
- Author Guidelines
- Graphical Abstracts
- Fabricating and Stating False Information
- Research Misconduct
- Post Publication Discussions and Corrections
- Publishing Ethics and Rectitude
- Increase Visibility of Your Article
- Archiving Policies
- Peer Review Workflow
- Order Your Article Before Print
- Promote Your Article
- Manuscript Transfer Facility
- Editorial Policies
- Allegations from Whistleblowers
- Announcements
Related Articles
-
Blood Serum Atherogenicity and Coronary Artery Calcification
Current Pharmaceutical Design Are Sex Differences in Outcomes of Patients with ACS from Observational Registries Similar to the Findings from Randomized Clinical Trials?
Current Pharmaceutical Design Modulators of the Vascular Endothelin Receptor in Blood Pressure Regulation and Hypertension
Current Molecular Pharmacology Editorial (Diabetes, Obesity and Vascular Disease - An Update)
Current Pharmaceutical Design Menopausal Hormone Therapy and Cardiovascular Risk: Where are we Now?
Current Vascular Pharmacology Transition of Care for Patients with Diabetes
Current Diabetes Reviews The Pathogenic Subpopulation of Th17 Cells in Obesity
Current Pharmaceutical Design Immunosuppressive and Antiviral Treatment of Inflammatory Cardiomyopathy
Recent Patents on Cardiovascular Drug Discovery Managing the Liabilities Arising from Structural Alerts: A Safe Philosophy for Medicinal Chemists
Current Medicinal Chemistry Thymosin β4 Protein Therapy for Cardiac Repair
Current Pharmaceutical Design Pharmacokinetics and Biotransformation of Tea Polyphenols
Current Drug Metabolism Hypoxia-Inducible Factor-1: A Molecular Hint of Physiological Changes in the Carotid Body During Long-Term Hypoxemia?
Current Drug Targets - Cardiovascular & Hematological Disorders Patents in the Diagnosis and Therapy of Neurocardiogenic Syncope
Recent Patents on Cardiovascular Drug Discovery Analytical Procedures for Secondary Metabolites Determination: Recent Trends and Future Perspectives
Letters in Drug Design & Discovery Prophylaxis of Erectile Function After Radical Prostatectomy with Phosphodiesterase Type 5 Inhibitors
Current Pharmaceutical Design Optimal Dosing Design for Antibiotic Therapy in the Elderly: A Pharmacokinetic and Pharmacodynamic Perspective
Recent Patents on Anti-Infective Drug Discovery The Relationship Between HIV Infection and Cardiovascular Disease
Current Cardiology Reviews Targeted Delivery of Anti-Inflammatory Agents to Tumors
Current Pharmaceutical Design Myxedema Heart Disease: A Rare Disease Entity: Case Report and Brief Review of the Literature
Current Hypertension Reviews The Long Pentraxin PTX3, a Soluble Pattern Recognition Receptor Involved in Innate Immunity,Inflammation and Female Fertility
Current Immunology Reviews (Discontinued)