Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Gene functional annotation by statistical analysis of biomedical articles.

T Theodosiou1, L Angelis, A Vakali

  • 1Department of Informatics, Aristotle University of Thessaloniki, Thessaloniki 54124, Greece. theodos@csd.auth.gr

International Journal of Medical Informatics
|June 20, 2006
PubMed
Summary

Linear Discriminant Analysis (LDA) offers a superior method for functional gene annotation compared to Support Vector Machines (SVM). This approach enhances the understanding of gene relationships and biological pathways through bio-ontology term assignment.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A comparative ultrastructural study of 'glue' production and secretion of the salivary glands in different species of theDrosophila melanogaster group.

Wilhelm Roux's archives of developmental biology·2017
Same author

The effect of sainfoin (Onobrychis viciifolia) and carob pods (Ceratonia siliqua) feeding regimes on the control of lamb coccidiosis.

Parasitology research·2016
Same author

Economic appraisal of the public control and prevention strategy against the 2010 West Nile Virus outbreak in Central Macedonia, Greece.

Public health·2015
Same author

Assessment of Chlamydia psittaci Shedding and Environmental Contamination as Potential Sources of Worker Exposure throughout the Mule Duck Breeding Process.

Applied and environmental microbiology·2015
Same author

Exploiting the temporal patterning of transient VEP signals: a statistical single-trial methodology with implications to brain-computer interfaces (BCIs).

Journal of neuroscience methods·2014
Same author

West Nile virus outbreak in humans, Greece, 2012: third consecutive year of local transmission.

Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin·2014

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Functional gene annotation is crucial for understanding gene relationships and biochemical pathways.
  • Bio-ontologies provide standardized vocabularies for describing gene functions.
  • Current methods for gene annotation involve data mining and machine learning techniques applied to biomedical literature.

Purpose of the Study:

  • To propose an alternative methodology for functional gene annotation.
  • To develop and validate classification models for gene annotation.
  • To explore data dimensionality reduction and graphical representations for result interpretation.

Main Methods:

  • Classification models were constructed using Linear Discriminant Analysis (LDA).
  • Model validation employed statistical analysis, hold-out samples, test datasets, and metrics (confusion matrix, accuracy, recall, precision, F-measure).

Related Experiment Videos

  • Graphical representations (boxplots, Andrew's curves, scatterplots) were used for result interpretation.

    • Main Results:

    • The LDA methodology was applied to a dataset for 12 Gene Ontology terms from biomedical articles.
    • LDA models demonstrated superior performance compared to Support Vector Machines (SVM).
    • LDA achieved a mean F-measure of 75.4%, outperforming SVM's mean F-measure of 68.7%.

    Conclusions:

    • Statistical methods, specifically LDA, are beneficial for functional gene annotation from biomedical text.
    • The proposed method offers good performance and provides interpretable results, offering insights into bio-text data structure.
    • LDA presents a viable and effective alternative to existing gene annotation techniques.