Abstract
Knowledge of mechanism of small molecules in metabolic pathway is critical to design specific and effective inhibitors for metabolic pathway. As some small molecules are involved in more than one pathway, it is crucial to use an accurate and robust approach to correctly map the small molecule in specific metabolic pathway that it is involved in. In this article, small molecules are studied using the Minimal-Redundancy-Maximal-Relevance-Forward Feature Search (mRMR-FFS) method combined with Multi-task learning method based on K-nearest neighbor (KNN) Algorithms method. Forty-five important chemical features were found based on 10-folds cross validation test from original data set containing 61 features. By applying KNN method with these forty-five selected features, the accuracy rate of prediction model could achieve 68.2% for the 10-folds cross validation test. It is promosing that our two stage scheme can be a useful approach for searching new effective competitive drugs in metabolic pathway.
Keywords: KEGG, KNN, metabolic pathway, mRMR, multi-task learning, small molecules.
Current Bioinformatics
Title:Small Molecules' Multi-Metabolic Pathways Prediction Using Physico- Chemical Features and Multi-Task Learning Method.
Volume: 8 Issue: 5
Author(s): Bing Niu, Lei Gu, Chunrong Peng, Juan Ding, Xiaochen Yuan and Wencong Lu
Affiliation:
Keywords: KEGG, KNN, metabolic pathway, mRMR, multi-task learning, small molecules.
Abstract: Knowledge of mechanism of small molecules in metabolic pathway is critical to design specific and effective inhibitors for metabolic pathway. As some small molecules are involved in more than one pathway, it is crucial to use an accurate and robust approach to correctly map the small molecule in specific metabolic pathway that it is involved in. In this article, small molecules are studied using the Minimal-Redundancy-Maximal-Relevance-Forward Feature Search (mRMR-FFS) method combined with Multi-task learning method based on K-nearest neighbor (KNN) Algorithms method. Forty-five important chemical features were found based on 10-folds cross validation test from original data set containing 61 features. By applying KNN method with these forty-five selected features, the accuracy rate of prediction model could achieve 68.2% for the 10-folds cross validation test. It is promosing that our two stage scheme can be a useful approach for searching new effective competitive drugs in metabolic pathway.
Export Options
About this article
Cite this article as:
Niu Bing, Gu Lei, Peng Chunrong, Ding Juan, Yuan Xiaochen and Lu Wencong, Small Molecules' Multi-Metabolic Pathways Prediction Using Physico- Chemical Features and Multi-Task Learning Method., Current Bioinformatics 2013; 8 (5) . https://dx.doi.org/10.2174/1574893611308050007
DOI https://dx.doi.org/10.2174/1574893611308050007 |
Print ISSN 1574-8936 |
Publisher Name Bentham Science Publisher |
Online ISSN 2212-392X |
- 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
Related Articles
-
Non-viral Gene Delivery and Therapeutics Targeting to Brain
Current Nanoscience A Study of Correlation between Anti-peroxidative Potential of Quercetin and Ascorbic Acid with Malondialdehyde by RP-HPLC
Current Chemical Biology Prostate Cancer Prevention in the Developing World - What are we Waiting for?
Current Pharmacogenomics and Personalized Medicine From Body Art to Anticancer Activities: Perspectives on Medicinal Properties of Henna
Current Drug Targets A Review on Skin Targeted Delivery of Bioactives as Ultradeformable Vesicles: Overcoming the Penetration Problem
Current Drug Targets Vitamin-D Receptor (VDR) Gene Polymorphisms (TaqI, FokI) in Turkish Patients with Hashimoto’s Thyroiditis: Relationship to the Levels of Vit-D and Cytokines
Endocrine, Metabolic & Immune Disorders - Drug Targets Atomic Layers in Electrochemical Biosensing Applications - Graphene and Beyond
Current Organic Chemistry The Suicidal Risks and Treatments, Seek Medications from Multi-disciplinary
Central Nervous System Agents in Medicinal Chemistry Implication of Medical Treatment for Surgical Strategies in IBD
Current Drug Targets Identification of Small Molecule Sulfonic Acids as Ecto-5'-Nucleotidase Inhibitors
Medicinal Chemistry Novel Therapeutic Targets in Neuropsychiatric Disorders: The Neuroepigenome
Current Pharmaceutical Design Cell-based Assays for Assessing Toxicity: A Basic Guide
Medicinal Chemistry Resveratrol and Cancer: Chemoprevention, Apoptosis, and Chemoimmunosensitizing Activities
Current Medicinal Chemistry - Anti-Cancer Agents Total Carotenoids Content and Assessment of Carotene Profile by High- Performance Liquid Chromatography in Selected Vegetables of Bangladesh with Special Reference to Some Unconventional Varieties
Current Chromatography Applications of Mannich Reaction in Total Syntheses of Natural Products
Current Organic Chemistry Development of a Quantitative Method for Monitoring 2-Mercaptobenzothiazole Based on Isotopic Iodoacetamide and Tandem MS
Current Analytical Chemistry Simultaneous Determination of Cytochrome P450 Oxidation Capacity in Humans: A Review on the Phenotyping Cocktail Approach
Current Pharmaceutical Biotechnology Synthesis of 4-(5-chloro-3-methyl-1-phenyl-1H-pyrazol-4-yl)-6-(substituted phenyl)pyrimidin-2-ol Analogues as Anti-Inflammatory and Analgesic Agents
Letters in Drug Design & Discovery Green Tea (Camellia sinensis) Protects Against Arsenic Neurotoxicity via Antioxidative Mechanism and Activation of Superoxide Dismutase Activity
Central Nervous System Agents in Medicinal Chemistry Coffea arabica Bean Extracts and Vitamin C: A Novel Combination Unleashes MCF-7 Cell Death
Current Pharmaceutical Biotechnology