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

Profiled support vector machines for antisense oligonucleotide efficacy prediction.

Gustavo Camps-Valls1, Alistair M Chalk, Antonio J Serrano-López

  • 1Grup de Processament Digital de Senyals, Universitat de València, Spain, C/ Dr, Moliner, 50, 46100 Burjassot, València, Spain. gustavo.camps@uv.es

BMC Bioinformatics
|September 24, 2004
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

Comprehensive analysis of the RBP regulome reveals functional modules and drug candidates in liver cancer.

Scientific reports·2026
Same author

GeneSNAKE: a Python package for simulation of gene regulatory networks and perturbation-induced expression data.

Bioinformatics advances·2026
Same author

Minute amounts of helicase-deficient truncated RECQL4 are sufficient for DNA replication.

EMBO reports·2026
Same author

KLHDC3 deficiency in mice reveals essential roles in development, survival, and adiposity via the DesCEND ubiquitin pathway.

BMC genomics·2026
Same author

Quest for Orthologs in the era of Data Deluge and AI: Challenges and Innovations in Orthology Prediction and Data Integration.

Journal of molecular evolution·2025
Same author

Minute amounts of helicase-deficient truncated RECQL4 are sufficient for DNA replication.

bioRxiv : the preprint server for biology·2025

Support Vector Machines (SVMs) effectively predict antisense oligonucleotide (AO) efficacy. A two-stage approach using SVM-based feature selection and profiled SVM prediction achieved superior accuracy compared to prior methods.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Machine Learning

Background:

  • Antisense oligonucleotide (AO) efficacy prediction is crucial for therapeutic development.
  • High-dimensional datasets with numerous features pose challenges for accurate modeling.
  • Support Vector Machines (SVMs) are suitable for complex prediction tasks like AO efficacy analysis.

Purpose of the Study:

  • To develop an optimal model for predicting antisense oligonucleotide (AO) efficacy.
  • To address feature selection challenges in high-dimensional AO datasets.
  • To compare standard and profiled SVM formulations for AO prediction.

Main Methods:

  • A two-stage strategy involving feature selection and prediction was employed.
  • Feature selection utilized correlation analysis, mutual information, and SVM-based recursive feature elimination (SVM-RFE).

Related Experiment Videos

  • Prediction involved standard and profiled SVM models, with profiled SVM assigning differential weights to training data.

    • Main Results:

    • SVM-RFE efficiently identified 14 key features related to energy and sequence motifs.
    • The profiled SVM achieved the best prediction performance (r = 0.44, ME = 0.022, RMSE = 0.278).
    • High and low efficacy AOs were predicted with success rates of 83.3% and 82.9%, respectively, outperforming previous methods.

    Conclusions:

    • SVMs provide a robust and accurate approach for AO efficacy prediction.
    • The profiled SVM formulation demonstrates superior performance over standard SVMs.
    • This methodology offers potential improvements for various prediction problems.