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

Mining protein function from text using term-based support vector machines.

Simon B Rice1, Goran Nenadic, Benjamin J Stapley

  • 1Faculty of Life Sciences, University of Manchester, UK. S.Rice@postgrad.manchester.ac.uk

BMC Bioinformatics
|June 18, 2005
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

Recognition and linking of discontinuous named entities in healthcare: a comparative performance analysis.

Frontiers in digital health·2026
Same author

Health inequalities in outpatient neurological conditions across a large UK urban population: a retrospective observational study using automated coding.

BMJ neurology open·2026
Same author

Development and validation of prediction models for predicting social care strengths and vulnerability in older people: Cohort study using routine data in Adult Social Care.

PloS one·2026
Same author

Towards a resource for multilingual lexicons: an MT assisted and human-in-the-loop multilingual parallel corpus with multi-word expression annotation.

Language resources and evaluation·2026
Same author

Extending BEHRT to UK Biobank: assessing transformer model performance in clinical prediction.

Frontiers in digital health·2026
Same author

A practical and nuanced framework for entity linking evaluation.

Journal of biomedical semantics·2026
Same journal

Covariance decomposition for distance based species tree estimation.

BMC bioinformatics·2026
Same journal

SNPio: a Python interface for population genomic data processing.

BMC bioinformatics·2026
Same journal

SpaHNR: a spatial domain identification method via sparse attention-based hierarchical node representation and multi-view contrastive learning.

BMC bioinformatics·2026
Same journal

OpenIMC: an open-source platform for analyzing single-cell and spatial proteomics by imaging mass cytometry.

BMC bioinformatics·2026
Same journal

NAP: an open source pipeline for cross-domain microbiome profiling using Nanopore sequencing-derived amplicon data.

BMC bioinformatics·2026
Same journal

SurvGME: an R package for survival analysis with graphical and measurement error models.

BMC bioinformatics·2026
See all related articles

This study used machine learning to assign Gene Ontology (GO) terms to human proteins from text. Performance was modest, highlighting the need for sufficient training data and relevant supporting documents for accurate protein function prediction.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Text mining is crucial in biology for extracting information.
  • The BioCreAtIvE exercise aimed to assess text mining system performance.
  • Task 2 focused on assigning Gene Ontology (GO) terms to human proteins and identifying supporting evidence.

Purpose of the Study:

  • To evaluate a supervised machine learning approach for assigning GO terms to human proteins.
  • To assess the system's ability to select relevant evidence from full-text documents.
  • To adapt document classification methods for biological text mining.

Main Methods:

  • Utilized a supervised machine learning approach, specifically support vector machines (SVMs).
  • Trained the model to assign protein function and identify supporting text passages.

Related Experiment Videos

  • Employed co-occurring terms extracted from documents as classification features.

    • Main Results:

    • Modest and variable performance in assigning GO terms and selecting relevant evidence (precision 3-50%).
    • The approach performed better with substantial relevant documents than with single or short texts.
    • Potential to mine protein annotations even without explicit statements linking proteins to GO terms.

    Conclusions:

    • Machine learning for protein function prediction from text requires sufficient training data and supporting evidence.
    • Combined document retrieval and GO term assignment show promise.
    • Integration of methods from BioCreAtIvE Task 1 and Task 2 is recommended.