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

Model selection methodology in supervised learning with evolutionary computation.

J J Rowland1

  • 1Department of Computer Science, University of Wales, Aberystwyth, SY23 3DB Wales, UK. jjr@aber.ac.uk

Bio Systems
|December 4, 2003
PubMed
Summary

Evolutionary methods offer powerful supervised learning in bioinformatics but risk overtraining. This study presents a computationally efficient model selection approach to ensure reliable bioinformatics model validation.

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

Primary and secondary metabolism, and post-translational protein modifications, as portrayed by proteomic analysis of Streptomyces coelicolor.

Molecular microbiology·2002
Same author

Discrimination of aerobic endospore-forming bacteria via electrospray-lonization mass spectrometry of whole cell suspensions.

Analytical chemistry·2001
Same author

A functional genomics strategy that uses metabolome data to reveal the phenotype of silent mutations.

Nature biotechnology·2001
Same author

Rapid analysis of high-dimensional bioprocesses using multivariate spectroscopies and advanced chemometrics.

Advances in biochemical engineering/biotechnology·1999
Same author

Genome-scale cloning and expression of individual open reading frames using topoisomerase I-mediated ligation.

Genome research·1999
Same author

Quantification of microbial productivity via multi-angle light scattering and supervised learning.

Biotechnology and bioengineering·1999

Area of Science:

  • Bioinformatics
  • Machine Learning
  • Computational Biology

Background:

  • Evolutionary methods are attractive for supervised learning in bioinformatics due to their expressive power and explicit models.
  • However, these methods are prone to overtraining and identifying spurious correlations in data.

Purpose of the Study:

  • To address the challenge of model selection in evolutionary algorithms for bioinformatics.
  • To present a computationally efficient approach for validating evolutionary models.

Main Methods:

  • Developed a novel model selection strategy tailored for evolutionary algorithms.
  • Applied the technique to two distinct bioinformatics datasets: metabolite determination and disease prediction from gene expression.

Main Results:

Related Experiment Videos

  • The proposed model selection method is effective and not excessively computationally intensive.
  • Demonstrated the technique's utility on real-world bioinformatics data.

Conclusions:

  • Appropriate model selection and validation are crucial for reliable evolutionary methods in bioinformatics.
  • The presented approach offers a practical solution for enhancing the trustworthiness of bioinformatics models derived from evolutionary computation.