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 bio-basis function neural network for protein peptide cleavage activity characterisation.

Zheng Rong Yang1, Jonathan Dry, Rebecca Thomson

  • 1Department of Computer Science, University of Exeter, Northcote House, The Queen's Drive, Exeter EX4 4QJ, UK. z.r.yang@ex.ac.uk

Neural Networks : the Official Journal of the International Neural Network Society
|February 16, 2006
PubMed
Summary

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

Global risk of selection and spread of Plasmodium falciparum histidine-rich protein 2 and 3 gene deletions.

Nature medicine·2025
Same author

Global risk of selection and spread of <i>Plasmodium falciparum</i> histidine-rich protein 2 and 3 gene deletions.

medRxiv : the preprint server for health sciences·2023
Same author

Lifestyle management in polycystic ovary syndrome - beyond diet and physical activity.

BMC endocrine disorders·2023
Same author

In silico prediction of Severe Acute Respiratory Syndrome Coronavirus 2 main protease cleavage sites.

Proteins·2021
Same author

Prevalence and factors associated with hypertension among people living with HIV/AIDS on antiretroviral therapy in Uganda.

The Pan African medical journal·2021
Same author

Prevalence of <i>Plasmodium falciparum</i> lacking histidine-rich proteins 2 and 3: a systematic review.

Bulletin of the World Health Organization·2020

A new bio-basis function neural network (BBFNN) analyzes protein peptides by treating amino acids as non-numerical data. This novel algorithm outperforms existing methods like multi-layer perceptrons and support vector machines.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Machine Learning

Background:

  • Protein peptides are fundamental to biological functions.
  • Analyzing peptide sequences with non-numerical attributes presents a challenge for traditional algorithms.
  • Existing neural networks like RBFNNs require numerical inputs, limiting their direct application to amino acid sequences.

Purpose of the Study:

  • To introduce a novel neural learning algorithm, the bio-basis function neural network (BBFNN), for analyzing protein peptides.
  • To adapt radial basis function neural networks (RBFNNs) for non-numerical attributes like amino acids.
  • To develop a method for characterizing protein peptide functional status based on amino acid composition.

Main Methods:

  • Developed the bio-basis function neural network (BBFNN) by replacing the radial basis function in RBFNNs with a bio-basis function.

Related Experiment Videos

  • Each basis in BBFNN is supported by a peptide, collectively forming a feature space where each basis represents a feature dimension.
  • Constructed a linear classifier in the feature space to characterize protein peptides by functional status, leveraging the principle that similar amino acid compositions indicate similar functions.

    • Main Results:

    • The BBFNN effectively utilizes the statistical significance of peptide similarity based on amino acid composition.
    • The proposed bio-basis function successfully encodes information from peptide data.
    • In two real-world case studies, BBFNN demonstrated superior performance compared to multi-layer perceptrons and support vector machines.

    Conclusions:

    • The BBFNN is a novel and effective neural learning algorithm for analyzing protein peptides with non-numerical attributes.
    • BBFNN offers an improved approach for peptide functional status characterization.
    • The algorithm shows significant potential for advancing bioinformatics and computational biology research.