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

Enhanced protein domain discovery by using language modeling techniques from speech recognition.

Lachlan Coin1, Alex Bateman, Richard Durbin

  • 1Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Cambridge CB10 1SA, United Kingdom.

Proceedings of the National Academy of Sciences of the United States of America
|April 2, 2003
PubMed
Summary

Statistical language models, adapted from speech recognition, significantly improve protein domain discovery. This method identified a novel Tf_Otx Pfam domain, offering insights into cone-rod dystrophy.

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

The Vertebrate Genomes Project Phase I: A global reference genome resource.

bioRxiv : the preprint server for biology·2026
Same author

Folding the unfoldable 2: using AlphaFold and ESMFold to explore spurious proteins.

Bioinformatics advances·2026
Same author

InterProScan 6: a modern large-scale protein function annotation pipeline.

Bioinformatics advances·2026
Same author

Viral non-coding RNA structure annotation and API-based data retrieval with Rfam and R2DT.

bioRxiv : the preprint server for biology·2026
Same author

A new phased assembly of the Antarctic spiny plunderfish provides novel insights into the evolution of the notothenioid radiation.

bioRxiv : the preprint server for biology·2026
Same author

Cryptic Diversity and Impacts of Domestication in the Black Soldier Fly (Hermetia illucens) Genome.

Genome biology and evolution·2026

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Speech recognition commonly employs probabilistic models like Hidden Markov Models (HMMs) for sound and word interpretation.
  • Similar probabilistic techniques have been applied to identify domains within protein sequences.
  • Language models enhance speech recognition by leveraging context to predict likely word combinations, improving accuracy.

Purpose of the Study:

  • To investigate the applicability of statistical language modeling techniques to protein domain discovery.
  • To enhance the accuracy and reliability of identifying functional domains in protein sequences.

Main Methods:

  • Adapted statistical language modeling methods, originally used in speech recognition, for protein sequence analysis.
  • Applied these models to identify conserved patterns and domains within amino acid sequences.

Related Experiment Videos

  • Utilized the enhanced domain recognition to analyze specific protein structures and mutations.

    • Main Results:

    • Demonstrated that statistical language modeling significantly improves protein domain recognition.
    • Successfully identified a previously unannotated Tf_Otx Pfam domain in the cone rod homeobox protein.
    • The discovered domain provides a potential mechanistic link between a specific mutation and disease.

    Conclusions:

    • Statistical language modeling is a powerful tool for advancing protein domain discovery in bioinformatics.
    • This approach offers a novel method for identifying functional regions and understanding protein evolution.
    • The findings shed light on the molecular basis of cone-rod dystrophy and suggest future research directions.