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

Analysis of molecular profile data using generative and discriminative methods.

E J Moler1, M L Chow, I S Mian

  • 1Department of Cell and Molecular Biology, Radiation Biology and Environmental Toxicology Group, Life Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California, USA.

Physiological Genomics
|December 20, 2000
PubMed
Summary

This study introduces a novel framework combining graphical models and Support Vector Machines (SVMs) to analyze molecular data. The approach effectively identifies cancer subtypes, potential biomarkers, and aids in diagnosis and prognosis.

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

Structural divergence in vertebrate phylogeny of a duplicated prototype galectin.

Genome biology and evolution·2014
Same author

Communication theory and multicellular biology.

Integrative biology : quantitative biosciences from nano to macro·2011
Same author

Cardiovascular disease, heart failure, chronic kidney disease and depression independently increase the risk of incident diabetes.

Diabetologia·2010
Same author

Statistical modeling of biomedical corpora: mining the Caenorhabditis Genetic Center Bibliography for genes related to life span.

BMC bioinformatics·2006
Same author

Ensemble attribute profile clustering: discovering and characterizing groups of genes with similar patterns of biological features.

BMC bioinformatics·2006
Same author

Robust sparse hyperplane classifiers: application to uncertain molecular profiling data.

Journal of computational biology : a journal of computational molecular cell biology·2005

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Understanding relationships between molecular profiles and domain knowledge is crucial for cancer research.
  • Existing methods may not fully leverage complex biological data for accurate classification and biomarker discovery.

Purpose of the Study:

  • To propose a modular framework integrating generative (graphical models) and discriminative (Support Vector Machines) methods.
  • To apply this framework to colon adenocarcinoma data for specimen classification and biomarker identification.
  • To develop a gene ranking and selection method based on model probability parameters.

Main Methods:

  • Utilized naive Bayes, graphical models, and Support Vector Machines (SVMs).
  • Applied unsupervised and supervised learning to transcription profile data from 62 colon adenocarcinoma specimens.

Related Experiment Videos

  • Developed a gene relevance ranking and selection method using class probability parameters.

    • Main Results:

    • Identified three distinct specimen classes (subtypes) and assigned tumor/nontumor labels.
    • Detected six potentially mislabeled specimens.
    • Achieved comparable or superior performance using only 50-200 top-ranked genes compared to all 1,988 genes.
    • Pinpointed approximately 90 marker genes for colon adenocarcinoma biology, therapeutics, and diagnostics.

    Conclusions:

    • The proposed framework effectively models molecular data and domain knowledge.
    • Identified potential biomarkers crucial for understanding cancer etiology and developing clinical tools.
    • Graphical models and SVMs show promise for decision support systems in oncology and biological network inference.