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

A robust meta-classification strategy for cancer detection from MS data.

Gyan Bhanot1, Gabriela Alexe, Babu Venkataraghavan

  • 1Center for Systems Biology, Institute for Advanced Study, Princeton, NJ, USA.

Proteomics
|December 13, 2005
PubMed
Summary

We developed a new method for identifying disease traits using noise reduction and machine learning. This approach successfully identified 11 protein biomarkers for prostate cancer, improving diagnostic accuracy.

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

CDK8 Inhibition Releases the Muscle Differentiation Block in Fusion-driven Alveolar Rhabdomyosarcoma.

Cancer discovery·2026
Same author

Splicing-associated network PAK1-CLK1/4-SRRM1 is a vulnerability to overcome chemoresistance in human and mouse acute myeloid leukemia.

Science translational medicine·2026
Same author

Distinct genomic and immunologic tumor evolution in germline TP53-driven breast cancers.

Nature communications·2026
Same author

Combining menin and MEK inhibition to target poor prognosis KMT2A-rearranged RAS pathway-mutant acute myeloid leukemia.

Blood advances·2026
Same author

Inhibition of USP7 Destabilizes the Noncanonical PRC1.1 Complex and Induces Neuroblastoma Differentiation.

Molecular cancer research : MCR·2026
Same author

The TP53 gene contains a diversity box that makes it more than a tumor suppressor.

Cell death and differentiation·2026

Area of Science:

  • Biomarker Discovery
  • Proteomics
  • Machine Learning Applications

Background:

  • Accurate phenotype identification is crucial for disease diagnosis and understanding.
  • Existing methods for analyzing complex proteomic data often face challenges with noise and require robust predictive models.
  • SELDI-TOF MS (Surface-Enhanced Laser Desorption/Ionization-Time of Flight Mass Spectrometry) generates high-dimensional data that necessitates advanced analytical techniques.

Purpose of the Study:

  • To propose and validate a novel, robust method for phenotype identification using stringent noise analysis and machine learning.
  • To apply the developed method to identify proteomic biomarkers for prostate cancer detection.
  • To assess the predictive performance of the proposed method in distinguishing cancer from non-cancer cases.

Main Methods:

Related Experiment Videos

  • Implementation of a stringent noise analysis and filtering procedure for proteomic data.
  • Integration of results from multiple machine learning tools to create a robust predictor.
  • Application and validation of the method on SELDI-TOF MS prostate cancer data.

Main Results:

  • Identification of 11 proteomic biomarkers associated with prostate cancer.
  • Achieved significantly improved prediction accuracy compared to previous analyses of the same dataset.
  • Demonstrated high diagnostic performance with 90.31% sensitivity and 98.81% specificity in distinguishing cancer from non-cancer cases.

Conclusions:

  • The proposed method offers a robust approach for phenotype identification through effective noise reduction and ensemble machine learning.
  • The identified proteomic biomarkers show potential for accurate prostate cancer diagnosis.
  • The methodology is generalizable to multi-phenotype prediction and can be applied to other data types, such as microarray data.