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

Recognition of adenosine triphosphate binding sites using parallel cascade system identification.

James R Green1, Michael J Korenberg, Robert David

  • 1Department of Electrical and Computer Engineering, Queen's University, Kingston, Ontario, Canada. james.green@ece.queensu.ca

Annals of Biomedical Engineering
|May 2, 2003
PubMed
Summary

Parallel cascade identification (PCI) effectively classifies proteins binding to ATP or GTP. Combining PCI with K-nearest-neighbor (KNN) classifiers significantly improves accuracy for protein binding prediction.

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

Spontaneous Dropwise Wicking for Scalable Integration of Salt-Loaded Nanoporous Sorbents.

ACS applied materials & interfaces·2026
Same author

Analytical Models to Optimize Tacrolimus Dosing in Solid Organ Transplantation: A Systematic Review.

Pharmaceutics·2026
Same author

Combining Fast Orthogonal Search with Deep Learning to Improve Low-Cost IMU Signal Accuracy.

Sensors (Basel, Switzerland)·2026
Same author

Sound and Vibration Exposure During Fixed-Wing Aircraft Neonatal Patient Transport.

Air medical journal·2026
Same author

Jigsaws soup: Improved laparoscopic surgical gesture classification, and skill level evaluation using model souping.

Computers in biology and medicine·2026
Same author

Leveraging learned representations and multitask learning for lysine methylation site discovery.

Scientific reports·2026

Area of Science:

  • Biochemistry
  • Bioinformatics
  • Computational Biology

Background:

  • Protein classification into structure/function families is crucial for understanding biological roles.
  • Identifying proteins that bind to specific molecules like ATP or GTP is essential for drug discovery and systems biology.
  • Existing methods for protein classification require refinement for improved accuracy and efficiency.

Purpose of the Study:

  • To apply Parallel Cascade Identification (PCI) for distinguishing between ATP/GTP-binding and non-binding proteins.
  • To evaluate the performance of PCI using different amino acid scales (Rose hydrophobicity and SARAH1).
  • To investigate the synergistic effect of combining PCI with K-nearest-neighbor (KNN) classifiers for enhanced protein binding prediction.

Main Methods:

Related Experiment Videos

  • Utilized Parallel Cascade Identification (PCI) to model protein sequence data.
  • Employed hydrophobicity scales (Rose et al. and SARAH1) for encoding protein sequences.
  • Developed Nearest-Neighbor and K-nearest-neighbor (KNN) classifiers.
  • Integrated PCI and KNN classifiers using quadratic discriminant analysis.

    • Main Results:

    • PCI achieved classification accuracies of 87.1% (Rose scale) and 88.8% (SARAH1 scale) via tenfold cross-validation.
    • KNN classifiers demonstrated accuracies of 88.0% and 90.8% on SARAH1-encoded data.
    • Combining PCI and KNN classifiers with quadratic discriminant analysis yielded a significant accuracy of 96.5% (twofold cross-validation on SARAH1 data).

    Conclusions:

    • PCI is a viable method for classifying proteins based on their binding affinities.
    • The SARAH1 scale enhances classification accuracy compared to the Rose hydrophobicity scale.
    • The combination of PCI and KNN classifiers offers a powerful approach for accurate prediction of protein-ligand interactions, specifically for ATP/GTP-binding proteins.