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

Regularized Least Squares Cancer classifiers from DNA microarray data.

Nicola Ancona1, Rosalia Maglietta, Annarita D'Addabbo

  • 1Istituto di Studi sui Sistemi Intelligenti per I'Automazione, CNR, Via Amendola 122/D-I, 70126 Bari, Italy. ancona@ba.issia.cnr.it

BMC Bioinformatics
|December 15, 2005
PubMed
Summary
This summary is machine-generated.

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

Benchmarking short- and long-read sequencing technologies for metagenomic profiling of microbiomes.

Scientific reports·2026
Same author

Federated, governed, and interoperable? The emerging architecture of public human genomic data infrastructures: a European perspective.

Frontiers in genetics·2026
Same author

Machine learning-based assessment of the healthy human gut mycobiota landscape using ITS1 DNA metabarcoding data.

BioData mining·2026
Same author

Observing abilities of satellite-tagged sea turtles: comparison of reconstructed temperature profiles with ocean model data in the Adriatic and Ionian Seas.

Scientific reports·2026
Same author

Enhanced Untargeted Metabolomics Based on High-Resolution Mass Spectrometry Reveals Global Rewiring Due to Mitochondrial Dysfunction in Yeast.

International journal of molecular sciences·2026
Same author

Single-cell transcriptomic profiling of human fetal neural stem cells isolated from the subventricular zone.

Frontiers in cell and developmental biology·2026

Regularized Least Squares (RLS) classifiers offer comparable performance to Support Vector Machines (SVM) for cancer classification using DNA microarrays. RLS classifiers present a simpler, computationally efficient alternative for analyzing gene expression data.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • DNA microarrays revolutionize cancer classification by enabling quantitative analysis.
  • Accurate cancer classification demands robust mathematical methods for predicting new cases from limited data.
  • This study compares Regularized Least Squares (RLS) classifiers with Support Vector Machines (SVM) for DNA microarray data analysis.

Purpose of the Study:

  • To assess the performance of RLS classifiers in cancer classification using DNA microarray data.
  • To compare RLS classifier performance against SVM, a leading technique in this field.
  • To investigate the impact of gene selection strategies and the number of genes on classifier performance.

Main Methods:

  • Evaluation of Regularized Least Squares (RLS) classifiers.

Related Experiment Videos

  • Comparison with Support Vector Machines (SVM) using Leave-One-Out (LOO) error.
  • Analysis across three distinct datasets, considering varying numbers of selected genes and different gene selection methods.

    • Main Results:

    • RLS classifiers demonstrate performance comparable to SVM classifiers, as evidenced by Leave-One-Out (LOO) error rates.
    • RLS machines solve classification problems using a linear system sized by features or training examples.
    • RLS machines provide an exact LOO error measure efficiently, requiring only one training iteration.

    Conclusions:

    • RLS classifiers are a viable alternative to SVM for cancer classification with gene expression data.
    • RLS classifiers offer advantages in simplicity and computational efficiency.
    • RLS classifiers exhibit comparable generalization abilities to SVM, even with limited gene expression data for new specimen classification.