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

Neural network models for promoter recognition.

A V Lukashin1, V V Anshelevich, B R Amirikyan

  • 1Institute of Molecular Genetics, USSR Academy of Sciences, Moscow.

Journal of Biomolecular Structure & Dynamics
|June 1, 1989
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

An earring for the double helix: assembly of topological links comprising duplex DNA and a circular oligodeoxynucleotide.

Journal of biomolecular structure & dynamics·2012
Same author

[Biophysics of the DNA molecule: a new trend].

Molekuliarnaia biologiia·2002
Same author

An artificial primosome: design, function, and applications.

Chembiochem : a European journal of chemical biology·2002
Same author

PNA beacons for duplex DNA.

Antisense & nucleic acid drug development·2001
Same author

High-purity preparation of a large DNA dumbbell.

Antisense & nucleic acid drug development·2001
Same author

Tuning DNA "strings": modulating the rate of DNA replication with mechanical tension.

Proceedings of the National Academy of Sciences of the United States of America·2001

This study uses neural networks to identify DNA promoter sites, achieving 94-99% accuracy. The model effectively distinguishes true promoters from random sequences.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Identifying promoter sites in DNA is crucial for understanding gene regulation.
  • Traditional methods face challenges in accurately and efficiently recognizing these regulatory sequences.
  • The need for advanced computational models is evident in genomics research.

Purpose of the Study:

  • To develop and evaluate a neural network model for accurate promoter site recognition in DNA sequences.
  • To estimate the maximum admissible network capacity for this specific bioinformatics problem.
  • To establish effective learning and testing protocols for the proposed neural network model.

Main Methods:

  • Utilized learning neural network models, specifically a block neural network architecture.

Related Experiment Videos

  • Estimated maximum network capacity based on experimental data of known promoter sequences.
  • Developed and applied learning and testing rules, using a small subset (approx. 10%) for training.

    • Main Results:

    • The neural network model achieved high promoter recognition efficiency, ranging from 94% to 99%.
    • The model successfully developed distinctive features ('keywords') for identifying promoters.
    • The probability of random sequences being misidentified as promoters was low, between 2% and 6%.

    Conclusions:

    • The developed block neural network model demonstrates a highly efficient and accurate method for promoter site recognition.
    • The model's ability to learn key features from a limited dataset highlights its robustness.
    • This approach offers a significant advancement in computational genomics for identifying regulatory DNA elements.