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 expression profile class prediction using linear Bayesian classifiers.

Musa H Asyali1

  • 1Department of Computer Engineering, Yasar University, Kazim Dirik Mah. 364 Sok. No: 5, Bornova 35500, Izmir, Turkey. musa.asyali@yasar.edu.tr

Computers in Biology and Medicine
|May 23, 2007
PubMed
Summary
This summary is machine-generated.

This study introduces a new linear Bayesian classifier for predicting tissue sample diagnostic categories using gene expression profiles. The method efficiently identifies key genes and achieves high accuracy, outperforming existing approaches.

Related Experiment Videos

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

Analysis of coronary angiography related psychophysiological responses.

Biomedical engineering online·2011
Same author

Sleep stage and obstructive apneaic epoch classification using single-lead ECG.

Biomedical engineering online·2010
Same author

Determining a continuous marker for sleep depth.

Computers in biology and medicine·2007
Same author

Feedback network controls photoreceptor output at the layer of first visual synapses in Drosophila.

The Journal of general physiology·2006
Same author

Segmentation of cDNA microarray spots using markov random field modeling.

Bioinformatics (Oxford, England)·2005
Same author

Use of meixner functions in estimation of Volterra kernels of nonlinear systems with delay.

IEEE transactions on bio-medical engineering·2005

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • DNA microarray technology enables accurate prediction of tissue sample diagnostic categories via gene expression profiles.
  • Existing gene expression profile classification methods have limitations.
  • Support Vector Machines (SVM) offer high prediction performance but lack interpretability and are complex.

Purpose of the Study:

  • To address shortcomings in current gene expression profile classification methods.
  • To propose a novel approach using linear Bayesian classifiers for improved diagnostic prediction.
  • To develop an efficient gene selection strategy for accurate classification.

Main Methods:

  • Constructing gene-level linear classifiers to identify genes with high prediction accuracy (low error rates).
  • Sequentially building a multi-dimensional linear classifier by incorporating the best-performing genes.
  • Minimizing prediction error by iteratively adding genes until optimal performance is reached.

Main Results:

  • The proposed linear Bayesian classifier approach outperforms Prediction Analysis of Microarrays (PAM).
  • The method achieves classification performance comparable to Support Vector Machines (SVM).
  • The gene selection scheme in PAM was found to be potentially misleading.

Conclusions:

  • The developed intuitive approach provides competitive classification performance for gene expression data.
  • This method offers an efficient way to identify biologically relevant genes for diagnostic prediction.
  • The approach balances predictive accuracy with interpretability, overcoming SVM's complexity and lack of insight.