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

Comparative experiments on learning information extractors for proteins and their interactions.

Razvan Bunescu1, Ruifang Ge, Rohit J Kate

  • 1Department of Computer Sciences, University of Texas, Austin, TX 78712, USA. razvan@cs.utexas.edu

Artificial Intelligence in Medicine
|April 7, 2005
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

Agnostic material classification using differential de Bruijn graphs of DNA imprints.

bioRxiv : the preprint server for biology·2026
Same author

Identifying membrane-bound transcriptional regulatory proteins from rare but evolutionarily conserved domain combinations.

Nucleic acids research·2026
Same author

A cross-vertebrate brain protein interaction map identifies conserved neural and non-neural complexes.

Cell reports·2026
Same author

A protein interactome for the last eukaryotic common ancestor illuminates the biochemical basis of modern genetic diseases.

Cell genomics·2026
Same author

Comparative proteomic profiling of receptor kinase signaling reveals key trafficking components enforcing plant stomatal development.

Science advances·2026
Same author

Validation and analysis of 12,000 AI-driven CAR-T designs in the <i>Bits to Binders</i> competitions.

bioRxiv : the preprint server for biology·2026

Machine learning effectively extracts human protein information and interactions from biomedical texts, improving upon previous methods. This automated approach aids in consolidating vast biological knowledge from databases like Medline.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Natural Language Processing

Background:

  • Biomedical text mining aims to consolidate biological knowledge.
  • Extracting human gene and protein data from Medline abstracts is challenging due to inconsistent naming conventions.
  • Automated information extraction systems are needed to process large volumes of biomedical literature.

Purpose of the Study:

  • To develop and evaluate machine learning systems for identifying human protein names in Medline abstracts.
  • To extract information on protein-protein interactions from biomedical text.
  • To assess the performance of various machine learning methods in biomedical information extraction.

Main Methods:

  • Utilized machine learning techniques, including support vector machines and maximum entropy, for protein name identification.

Related Experiment Videos

  • Employed rule induction methods for extracting protein interaction data.
  • Trained and tested systems on a manually annotated corpus of approximately 1000 Medline abstracts.

    • Main Results:

    • Machine learning approaches achieved higher accuracy in identifying human proteins compared to prior methods.
    • Rule induction methods demonstrated superior precision in identifying protein interactions over manually created rules.
    • Demonstrated the effectiveness of automated systems in extracting complex biological information.

    Conclusions:

    • Machine learning shows significant promise for automatically building robust information extraction systems for biomedical text.
    • The study provides a comparative analysis of various methods' strengths on a substantial human-annotated dataset.
    • Automated extraction of protein information and interactions can accelerate biological knowledge discovery.