Abstract
Recent times have seen flooding of biological data into the scientific community. Due to increase in large amounts of data from genome and other sequencing projects become available, being diverted on to Insilco approach for data collection and prediction has become a priority also progresses in sequencing technologies have found an exponential function rise in the number of newly found enzymes. Commonly, function of such enzymes is determined by experiments that can be time consuming and costly. As new approaches are needed to determine the functions of the proteins these genes encode. The protein parameters that can be used for an enzyme/ non-enzyme classification includes features of sequences like amino acid composition, dipeptide composition, grand Average of hydropathicity (GRAVY), probability of being in alpha helix, probability of being in beta sheet Probability of being in a turn. We show how large-scale computational analysis can help to address this challenge by help of java and support vector machine library. In this paper, a recently developed machine learning algorithm referred to as the svm library Learning Machine is used to classify protein sequences with six main classes of enzyme data downloaded from a public domain database. Comparative studies on different type of kernel methods like 1.radial basis function, 2.polynomial available in SVM library. Results show that RBF method take less time in training and give more accurate result then other kernel methods to also less training time compared to other kernel methods. The classification accuracy of RBF is also higher than various methods in respect of available sequences data.
Keywords: Enzyme class, protein, support vector machine, kernel, java, functional class, model.
Current Topics in Medicinal Chemistry
Title:ENZPRED-Enzymatic Protein Class Predicting by Machine Learning
Volume: 13 Issue: 14
Author(s): Kirtan Dave and Hetalkumar Panchal
Affiliation:
Keywords: Enzyme class, protein, support vector machine, kernel, java, functional class, model.
Abstract: Recent times have seen flooding of biological data into the scientific community. Due to increase in large amounts of data from genome and other sequencing projects become available, being diverted on to Insilco approach for data collection and prediction has become a priority also progresses in sequencing technologies have found an exponential function rise in the number of newly found enzymes. Commonly, function of such enzymes is determined by experiments that can be time consuming and costly. As new approaches are needed to determine the functions of the proteins these genes encode. The protein parameters that can be used for an enzyme/ non-enzyme classification includes features of sequences like amino acid composition, dipeptide composition, grand Average of hydropathicity (GRAVY), probability of being in alpha helix, probability of being in beta sheet Probability of being in a turn. We show how large-scale computational analysis can help to address this challenge by help of java and support vector machine library. In this paper, a recently developed machine learning algorithm referred to as the svm library Learning Machine is used to classify protein sequences with six main classes of enzyme data downloaded from a public domain database. Comparative studies on different type of kernel methods like 1.radial basis function, 2.polynomial available in SVM library. Results show that RBF method take less time in training and give more accurate result then other kernel methods to also less training time compared to other kernel methods. The classification accuracy of RBF is also higher than various methods in respect of available sequences data.
Export Options
About this article
Cite this article as:
Dave Kirtan and Panchal Hetalkumar, ENZPRED-Enzymatic Protein Class Predicting by Machine Learning, Current Topics in Medicinal Chemistry 2013; 13 (14) . https://dx.doi.org/10.2174/15680266113139990118
DOI https://dx.doi.org/10.2174/15680266113139990118 |
Print ISSN 1568-0266 |
Publisher Name Bentham Science Publisher |
Online ISSN 1873-4294 |
Call for Papers in Thematic Issues
Adaptogens—History and Future Perspectives
Adaptogens are pharmacologically active compounds or plant extracts that are associated with the ability to enhance the body’s stability against stress. The intake of adaptogens is associated not only with a better ability to adapt to stress and maintain or normalise metabolic functions but also with better mental and physical ...read more
Addressing the Most Common Causes of Death with Niacin/NAD and Inositol Polyphosphates
The most common causes of death in the world are cardiovascular disease (CVD) and cancer. These are perhaps best addressed by reducing lipodystrophy and blockages with niacin and inositol polyphosphates (e.g., IP6+inositol) respectively when addressing CVD. Niacin serves as a vitamin by virtue of its role as a skeletal precursor ...read more
AlphaFold in Medicinal Chemistry: Opportunities and Challenges
AlphaFold, a groundbreaking AI tool for protein structure prediction, is revolutionizing drug discovery. Its near-atomic accuracy unlocks new avenues for designing targeted drugs and performing efficient virtual screening. However, AlphaFold's static predictions lack the dynamic nature of proteins, crucial for understanding drug action. This is especially true for multi-domain proteins, ...read more
Artificial intelligence for Natural Products Discovery and Development
Our approach involves using computational methods to predict the potential therapeutic benefits of natural products by considering factors such as drug structure, targets, and interactions. We also employ multitarget analysis to understand the role of drug targets in disease pathways. We advocate for the use of artificial intelligence in predicting ...read more

- Author Guidelines
- Bentham Author Support Services (BASS)
- 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
-
Novel Kynurenic Acid Analogues in the Treatment of Migraine and Neurodegenerative Disorders: Preclinical Studies and Pharmaceutical Design
Current Pharmaceutical Design Potential Options to Treat Hypertriglyceridaemia
Current Drug Targets A Glance at Taenia Saginata Infection, Diagnosis, Vaccine, Biological Control and Treatment
Infectious Disorders - Drug Targets WISH Cells as a Model for the “In Vitro” Study of Amnion Pathophysiology
Current Drug Targets - Immune, Endocrine & Metabolic Disorders Novel Concept in the Mechanism of Injury and Protection of Gastric Mucosa: Role of Renin-Angiotensin System and Active Metabolites of Angiotensin
Current Medicinal Chemistry Genetics of Cancer Susceptibility
Current Genomics Enhancement of Dissolution Profile of Poorly Water Soluble Drugs by Using Techniques of Nanocrystallization
Current Nanomedicine Role of the Hsp90-Immunophilin Heterocomplex in Cancer Biology
Current Cancer Therapy Reviews Therapeutic Perspectives of Drugs Targeting Toll-Like Receptors Based on Immune Physiopathology Theory of Alzheimers Disease
CNS & Neurological Disorders - Drug Targets Inflammatory Cytokines and Atherosclerosis Possible Application for the Gene Therapy
Vascular Disease Prevention (Discontinued) Anticancer Activity of the Ascidian Polyclinum Indicum against Cervical Cancer Cells (HeLa) Mediated through Apoptosis Induction
Medicinal Chemistry The Redox Biology of Schistosome Parasites and Applications for Drug Development
Current Pharmaceutical Design Editorial [Hot Topic: Anti-Neuroinflammatory Effects of Psychopharmaceuticals: Further than Monoamine Modulators (Guest Editor: Sadayuki Hashioka)]
Mini-Reviews in Medicinal Chemistry Shared Signaling Pathways Between Endocrine and Immune System Receptors: The Model of Gamma Chain
Current Signal Transduction Therapy Target Fishing of Calactin, Calotropin and Calotoxin Using Reverse Pharmacophore Screening and Consensus Inverse Docking Approach
Current Drug Discovery Technologies Deregulation of Apoptosis - Is it Still an Important Issue in Pathogenesis of Chronic Lymphocytic Leukemia?
Current Cancer Drug Targets Different Sides of Depression in the Elderly: An In-depth View on the Role of Aβ Peptides
Current Medicinal Chemistry Synaptic Inputs of Neural Afferent Pathways to Vasopressin- and Oxytocin-Secreting Neurons of Supraoptic and Paraventricular Hypothalamic Nuclei
Endocrine, Metabolic & Immune Disorders - Drug Targets Polypharmacology of Natural Products Harnessed from the Ocean
Mini-Reviews in Organic Chemistry Terpenoids: Sources, Structure Elucidation and Therapeutic Potential in Inflammation
Current Topics in Medicinal Chemistry