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

Prediction of virus-host infectious association by supervised learning methods.

Mengge Zhang1, Lianping Yang2, Jie Ren1

  • 1Molecular and Computational Biology Program, University of Southern California, Los Angeles, California, USA.

BMC Bioinformatics
|April 1, 2017
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

In Situ Bacterioplankton Growth Partitioning by High-Resolution Metatranscriptomics.

Environmental microbiology reports·2026
Same author

The dynamic mesoscale sink and source niches for eukaryotic phytoplankton in a subtropical gyre.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

Converting relative amplicon abundances to absolute abundances via flow cytometry: metagenomic validation and application to long ocean transects.

ISME communications·2026
Same author

A multidomain lens on the temporal dynamics of surface microbial communities in the Southern Ocean (2013-2019).

Marine environmental research·2026
Same author

Metagenomic Hi‑C Protocols for Viral Genome Binning, Taxonomic Annotation, and Interaction Network Visualization.

Current protocols·2026
Same author

Benchmarking alignment strategies for Hi-C reads in metagenomic Hi-C data.

Genome biology·2026

Machine learning, specifically random forest, accurately predicts virus-host associations using viral sequence word frequencies. This method enhances understanding of microbial communities and aids in metagenomic analysis.

Area of Science:

  • Microbiology
  • Bioinformatics
  • Computational Biology

Background:

  • Understanding virus-host interactions is crucial for microbial community dynamics.
  • Metagenomic data often yields viral contigs lacking host information.
  • Existing methods for virus-host association are limited.

Purpose of the Study:

  • To investigate the efficacy of machine learning methods for predicting virus-host infectious associations.
  • To develop a robust approach for identifying viral hosts from sequence data.

Main Methods:

  • Utilized four word frequency representations of viral sequences.
  • Applied five machine learning algorithms (logistic regression, SVM, random forest, Gaussian NB, Bernoulli NB).
  • Evaluated performance using Area Under the Curve (AUC) and developed a maximum likelihood method for viral tagging experiments.

Related Experiment Videos

Main Results:

  • Random forest with relative word frequency achieved high accuracy (AUC > 0.85 for all nine genera, >0.98 for five).
  • The method accurately predicted hosts for viral contigs >= 1kbps in metagenomic data.
  • Outperformed existing methods based on Manhattan and [Formula: see text] dissimilarity measures.

Conclusions:

  • Random forest with relative word frequencies effectively predicts viruses and viral contigs for specific bacterial hosts.
  • A maximum likelihood approach can estimate infectious virus fractions in viral tagging experiments.