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

Protein structure prediction: selecting salient features from large candidate pools

K J Cherkauer1, J W Shavlik

  • 1Computer Sciences Department, University of Wisconsin-Madison 53706, USA.

Proceedings. International Conference on Intelligent Systems for Molecular Biology
|January 1, 1993
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

Biomedical informatics training at the University of Wisconsin-Madison.

Yearbook of medical informatics·2007
Same author

Evaluating machine learning approaches for aiding probe selection for gene-expression arrays.

Bioinformatics (Oxford, England)·2002
Same author

Neural network input representations that produce accurate consensus sequences from DNA fragment assemblies.

Bioinformatics (Oxford, England)·1999
Same author

Understanding time series networks: a case study in rule extraction.

International journal of neural systems·1997
Same author

Increasing consensus accuracy in DNA fragment assemblies by incorporating fluorescent trace representations.

Proceedings. International Conference on Intelligent Systems for Molecular Biology·1997
Same author

Improving the quality of automatic DNA sequence assembly using fluorescent trace-data classifications.

Proceedings. International Conference on Intelligent Systems for Molecular Biology·1996
Same journal

Proceedings of the 8th International Conference on Intelligent Systems for Molecular Biology (ISMB 2000). San Diego, California, USA. August 19-23, 2000.

Proceedings. International Conference on Intelligent Systems for Molecular Biology·2001
Same journal

Analysis of gene expression data with pathway scores.

Proceedings. International Conference on Intelligent Systems for Molecular Biology·2000
Same journal

Towards a complete map of the protein space based on a unified sequence and structure analysis of all known proteins.

Proceedings. International Conference on Intelligent Systems for Molecular Biology·2000
Same journal

Mining for putative regulatory elements in the yeast genome using gene expression data.

Proceedings. International Conference on Intelligent Systems for Molecular Biology·2000
Same journal

A multiple alignment algorithm for metabolic pathway analysis using enzyme hierarchy.

Proceedings. International Conference on Intelligent Systems for Molecular Biology·2000
Same journal

Sequence database search using jumping alignments.

Proceedings. International Conference on Intelligent Systems for Molecular Biology·2000
See all related articles

We developed DT-SELECT, a fast feature selection method for protein secondary structure prediction. It efficiently identifies crucial features from large datasets, though performance gains were minimal when augmenting neural networks.

Area of Science:

  • Bioinformatics
  • Machine Learning
  • Computational Biology

Background:

  • Predicting protein secondary structure is crucial for understanding protein function.
  • Inductive learning algorithms require effective feature sets for accurate predictions.
  • Large feature pools present challenges in selecting relevant and non-redundant features.

Purpose of the Study:

  • To introduce DT-SELECT, a parallel approach for efficient feature selection.
  • To enable rapid selection of small, non-redundant feature sets from extensive pools.
  • To assess the utility of DT-SELECT-chosen features in protein secondary structure prediction.

Main Methods:

  • DT-SELECT builds a decision tree using features from a large pool to classify training examples.

Related Experiment Videos

  • Selected features from the decision tree provide a compact data description.
  • These features are then used to augment standard artificial neural network representations.

    • Main Results:

    • DT-SELECT rapidly selects small, non-redundant feature sets from pools of hundreds of thousands of features.
    • Augmenting artificial neural networks with DT-SELECT-chosen features yielded minimal performance improvements.
    • This outcome was observed even when selecting features from very large feature pools.

    Conclusions:

    • DT-SELECT is an efficient method for feature selection in bioinformatics tasks.
    • The limited performance gains suggest that feature set size or complexity may not be the sole determinants of prediction accuracy.
    • Further investigation is needed to understand the reasons behind the minimal performance enhancement.