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

An efficient algorithm for detecting frequent subgraphs in biological networks.

Mehmet Koyutürk1, Ananth Grama, Wojciech Szpankowski

  • 1Department of Computer Sciences, Purdue University, West Lafayette, IN 47907, USA. koyuturk@cs.purdue.edu

Bioinformatics (Oxford, England)
|July 21, 2004
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

EV-Planner: a machine learning approach to electric vehicle charging infrastructure planning.

Frontiers in artificial intelligence·2026
Same author

Learning to take it personally: Precision drug repurposing through patient-specific loss on knowledge graphs using Biobank data.

Journal of biomedical informatics·2026
Same author

scProfiterole: Clustering of Single-Cell Proteomic Data Using Graph Contrastive Learning via Spectral Filters.

bioRxiv : the preprint server for biology·2026
Same author

What Does Next-Generation Mass Spectrometry Offer for Proteomics? A Comprehensive Platform Comparison.

Journal of proteome research·2026
Same author

Interpretable Machine Learning to Identify Risk Factors for Recidivism in Intimate Partner Violence.

AMIA ... Annual Symposium proceedings. AMIA Symposium·2026
Same author

Redox Stoichiometry at the Single-Residue Level Using Mass Spectrometry Reveals Dynamic Methionine Sulfoxide Speciation in Actin and Calmodulin during Brain Aging.

Journal of proteome research·2025

This study introduces a novel algorithm for identifying recurring patterns in biological networks. The new method simplifies graph analysis, making complex network pattern detection computationally feasible and fast, even for large datasets like metabolic pathways.

Area of Science:

  • Computational Biology
  • Bioinformatics
  • Network Analysis

Background:

  • Analyzing the vast amount of network and interaction data in molecular biology presents significant computational challenges.
  • Traditional graph theoretic approaches for biological network analysis are often computationally intractable due to their complexity, similar to subgraph isomorphism problems.

Purpose of the Study:

  • To develop an innovative algorithm for the efficient detection of frequently occurring patterns and modules within biological networks.
  • To address the computational hardness associated with analyzing large-scale biological network data.

Main Methods:

  • Introduction of a novel graph simplification technique specifically designed for biological networks.
  • Development of an algorithm that leverages this simplification to make pattern detection computationally tractable.

Related Experiment Videos

Main Results:

  • The proposed algorithm successfully detects frequently occurring patterns in biological networks.
  • Experimental results demonstrate rapid extraction of patterns from metabolic pathways (KEGG database) within seconds.
  • The algorithm's efficiency significantly improves upon existing methods for biological network analysis.

Conclusions:

  • The developed algorithm and model offer a computationally tractable solution for analyzing biological networks.
  • The approach is versatile and can be applied to various types of biological networks with minimal modifications.
  • Open-source implementation is available, facilitating wider adoption and research in the field.