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

Class prediction and discovery using gene microarray and proteomics mass spectroscopy data: curses, caveats,

R L Somorjai1, B Dolenko, R Baumgartner

  • 1Institute for Biodiagnostics, National Research Council Canada, Winnipeg, MB, Canada R3B 1Y6. Ray.Somorjai@nrc-cnrc.gc.ca

Bioinformatics (Oxford, England)
|August 13, 2003
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

Percutaneous sacroiliac screw fixation with a 3D robot-assisted image-guided navigation system : Technical solutions.

Operative Orthopadie und Traumatologie·2024
Same author

Conservative or operative therapy in patients with a fragility fracture of the pelvis: study protocol for a prospective, randomized controlled trial.

Trials·2024
Same author

Investigating Stability in Subgroup Identification for Stratified Medicine.

Pharmaceutical statistics·2024
Same author

CRISPR-Cas12a nucleases function with structurally engineered crRNAs: SynThetic trAcrRNA.

Scientific reports·2022
Same author

A perspective on the role of uncertainty in sustainability science and engineering.

Resources, conservation, and recycling·2020
Same author

Evidence that a deviation in the kynurenine pathway aggravates atherosclerotic disease in humans.

Journal of internal medicine·2020
Same journal

Cross-Domain Transfer Learning from Peptides to Metabolites Using a Multi-Property Fine-Tuned LLM.

Bioinformatics (Oxford, England)·2026
Same journal

Biomedical Concept Recognition with Error-aware Negative-enhanced Ranking Framework.

Bioinformatics (Oxford, England)·2026
Same journal

TEDLH: Domain HMMs for sensitive detection of remote homologues.

Bioinformatics (Oxford, England)·2026
Same journal

PLNFGL: Joint Estimation of Multi-Condition Gene Networks from Single-cell RNA-seq Data.

Bioinformatics (Oxford, England)·2026
Same journal

MCFST: Spatial domain identification method based on multi-view graph convolutional network and graph fusion network.

Bioinformatics (Oxford, England)·2026
Same journal

SpaBiT: Enhancing Spatial Transcriptomics Resolution via Bidirectional Attention Transformers.

Bioinformatics (Oxford, England)·2026
See all related articles

High-dimensional data analysis, common in genomics and proteomics, faces challenges from the curse of dimensionality and dataset sparsity. These issues can make disease classification results statistically unreliable and difficult to interpret.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Data Science

Background:

  • Analysis of high-dimensional biomedical data (microarrays, proteomics, spectroscopy) is constrained by the curse of dimensionality and dataset sparsity.
  • These challenges significantly impact the reliability of disease classification models.

Purpose of the Study:

  • To investigate the influence of dimensionality and sparsity on disease classification using real-world biomedical datasets.
  • To evaluate the statistical validity and biological relevance of classification results derived from sparse datasets.

Main Methods:

  • Utilized publicly available microarray and proteomics datasets.
  • Employed simple classification algorithms to assess outcomes under dimensionality and sparsity constraints.
  • Examined feature extraction/reduction methods and their impact on sample-to-feature ratios.

Related Experiment Videos

Main Results:

  • Dataset sparsity can render classification outcomes statistically suspect, even with feature reduction.
  • Multiple 'optimal' feature sets can yield perfect classification, leading to interpretational challenges and questioning biological relevance.
  • Identified issues with non-unique optimal feature sets in sparse data.

Conclusions:

  • The curse of dimensionality and dataset sparsity pose significant challenges for reliable disease classification in biomedical data analysis.
  • Suggests an approach to evaluate and compare the quality of classifiers derived from sparse datasets.
  • Highlights the need for careful interpretation of classification results and feature sets in high-dimensional, low-sample data.