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

Extracting sequence features to predict protein-DNA interactions: a comparative study.

Qing Zhou1, Jun S Liu

  • 1Department of Statistics, University of California, Los Angeles, CA 90095, USA. zhou@stat.ucla.edu

Nucleic Acids Research
|June 17, 2008
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

JADE: Joint Alignment and Deep Embedding for Multi-Slice Spatial Transcriptomics.

Advances in neural information processing systems·2026
Same author

Phenotypic prediction of missense variants via deep contrastive learning.

Nature biomedical engineering·2026
Same author

DEDUCE: statistical inference on disease-associated genes uncovers tissue-disease associations.

NAR genomics and bioinformatics·2026
Same author

Designing strongly coupled polaritonic structures via statistical machine learning.

Proceedings of the National Academy of Sciences of the United States of America·2025
Same author

JADE: Joint Alignment and Deep Embedding for Multi-Slice Spatial Transcriptomics.

bioRxiv : the preprint server for biology·2025
Same author

Participation bias in the estimation of heritability and genetic correlation.

Proceedings of the National Academy of Sciences of the United States of America·2025
Same journal

Correction to 'New origin firing is inhibited by APC/CCdh1 activation in S-phase after severe replication stress'.

Nucleic acids research·2026
Same journal

VeloRM: disentangling pre- and post-splicing RNA modification dynamics at single-cell resolution.

Nucleic acids research·2026
Same journal

Accessibility of telomeric overhangs to stabilizing small-molecule ligands.

Nucleic acids research·2026
Same journal

Multivalent interactions mediate SNAIL transcription factor stimulation of the nucleosome deacetylase activity of the CoREST complex.

Nucleic acids research·2026
Same journal

Genome-wide mapping of DNA G-quadruplexes in Trypanosoma brucei chromatin reveals enrichment in coding regions and transcription start sites.

Nucleic acids research·2026
Same journal

Correction to 'The Gene Ontology knowledgebase in 2026'.

Nucleic acids research·2026
See all related articles

Predicting protein-DNA interactions is crucial. Advanced machine learning models, like BART and boosting, outperform traditional methods for predicting transcription factor binding sites, revealing complex interactions.

Area of Science:

  • Genomics
  • Computational Biology
  • Molecular Biology

Background:

  • Predicting transcription factor (TF) binding to DNA is essential for understanding gene regulation.
  • Traditional methods like position-specific weight matrices (PWMs) have limitations in accuracy and feature identification.

Purpose of the Study:

  • To systematically evaluate advanced predictive modeling approaches for TF-DNA binding.
  • To compare the performance of various machine learning methods against PWMs.

Main Methods:

  • Applied regression frameworks to infer relationships between genomic sequences and gene expression/binding intensities.
  • Examined stepwise linear regression, MARS, neural networks, SVM, boosting, and BART.
  • Utilized simulated and whole-genome ChIP-chip datasets for Oct4 and Sox2 TFs in human embryonic stem cells.

Related Experiment Videos

Main Results:

  • Predictive modeling approaches significantly improved prediction accuracy compared to PWMs.
  • Identified biologically relevant features, including TF-TF interactions.
  • BART and boosting demonstrated the most robust and superior performance.

Conclusions:

  • Advanced machine learning, particularly BART and boosting, offers enhanced predictive power for TF-DNA interactions.
  • These methods facilitate the discovery of complex regulatory mechanisms beyond simple sequence motifs.