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

A non-parametric model for transcription factor binding sites.

Oliver D King1, Frederick P Roth

  • 1Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, 250 Longwood Avenue, SGMB-322, Boston, MA 02115, USA.

Nucleic Acids Research
|September 23, 2003
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

Image-based, pooled phenotyping reveals multidimensional, disease-specific variant effects.

Cell·2026
Same author

Global genetic interaction network of a human cell maps conserved principles and informs functional interpretation of gene co-essentiality profiles.

Cell·2026
Same author

Experimental assessment of AI-based interactome mapping.

Nature communications·2026
Same author

Selective chr21 homolog silencing reveals polymorphisms influence the epigenetic silencing and functional dosage of RWDD2B.

American journal of human genetics·2026
Same author

A scalable approach to resolving variants of uncertain significance.

bioRxiv : the preprint server for biology·2026
Same author

Mapping the Functional Landscape of <i>KCNQ1</i> to Define Ion Channel Mechanisms and Arrhythmia Risk.

medRxiv : the preprint server for health sciences·2025

We developed a flexible non-parametric model for transcription factor binding sites that outperforms standard position-specific scoring matrices (PSSMs). This new method improves the prediction of transcription factor binding sites, enhancing genomic analysis.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Transcription factor binding sites (TFBS) are crucial for gene regulation.
  • Position-specific scoring matrices (PSSMs) are commonly used to represent TFBS but have limitations in modeling complex dependencies.
  • Accurate TFBS identification is essential for understanding gene expression and disease mechanisms.

Purpose of the Study:

  • To introduce a novel non-parametric representation for transcription factor binding sites.
  • To demonstrate that this representation can model arbitrary dependencies between positions within binding sites.
  • To evaluate the performance of this new representation against standard PSSMs in predicting unseen binding sites.

Main Methods:

  • Developed a non-parametric model for TFBS representation.

Related Experiment Videos

  • The model smoothly interpolates between empirical binding site distributions and PSSMs by varying two parameters.
  • Evaluated model performance using 10-fold cross-validation on known TFBS for 95 transcription factors from TRANSFAC.

    • Main Results:

    • The non-parametric representation outperformed PSSMs on 65 to 89 out of 95 transcription factors.
    • Performance varied based on the selection of the two adjustable parameters.
    • The study explored incorporating this representation into frameworks for identifying TFBS in unaligned promoter regions.

    Conclusions:

    • The proposed non-parametric representation offers a more flexible and accurate method for modeling transcription factor binding sites compared to PSSMs.
    • This approach enhances the ability to identify regulatory elements in genomic sequences.
    • The findings suggest potential improvements for computational tools used in genomic analysis and disease research.