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

Generalized Hierarchical Sparse Model for Arbitrary-Order Interactive Antigenic Sites Identification in Flu Virus

Lei Han1, Yu Zhang2, Xiu-Feng Wan3

  • 1Department of Statistics, Rutgers University.

KDD : Proceedings. International Conference on Knowledge Discovery & Data Mining
|April 11, 2017
PubMed
Summary

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

Effects of miR-200a and FH535 combined with taxol on proliferation and invasion of gastric cancer.

Pathology, research and practice·2018
Same author

A Dual-Functional Luminescent MOF Sensor for Phenylmethanol Molecule and Tb<sup>3+</sup> Cation.

Inorganic chemistry·2018
Same author

The Development and Test of a Sensor for Measurement of the Working Level of Gas-Liquid Two-Phase Flow in a Coalbed Methane Wellbore Annulus.

Sensors (Basel, Switzerland)·2018
Same author

A Thioflavin T-induced G-Quadruplex Fluorescent Biosensor for Target DNA Detection.

Analytical sciences : the international journal of the Japan Society for Analytical Chemistry·2018
Same author

A metal-organic framework derived hierarchical nickel-cobalt sulfide nanosheet array on Ni foam with enhanced electrochemical performance for supercapacitors.

Dalton transactions (Cambridge, England : 2003)·2018
Same author

Functional analysis of a type 2C protein phosphatase gene from Ammopiptanthus mongolicus.

Gene·2018

We introduce a generalized hierarchical sparse model (GHSM) to effectively analyze high-order feature interactions in biological data. This method addresses the challenge of high dimensionality, improving interpretability and identifying complex covariate patterns.

Area of Science:

  • Statistical modeling
  • Bioinformatics
  • Machine learning

Background:

  • Regression models with feature interactions enhance biological data interpretability.
  • High-order interactions lead to exponentially expanded feature spaces, posing a dimensionality challenge.
  • Existing hierarchical sparse models (HSM) are limited to pairwise interactions.

Purpose of the Study:

  • To generalize hierarchical sparse models (HSM) for analyzing arbitrary-order feature interactions.
  • To develop a robust method for identifying complex synergistic covariate patterns in biological data.
  • To overcome the limitations of existing methods in discovering important high-order interactions.

Main Methods:

  • Proposed a generalized hierarchical sparse model (GHSM) with an L1 penalty and heredity constraints.
Keywords:
Analytical SolutionAntigenic Sites IdentificationHeredity StructureHierarchical SparsityHigh-Order Interaction

Related Experiment Videos

  • Developed an efficient optimization algorithm using GIST and ADMM to solve the non-convex objective function.
  • Decoupled variables into three subproblems with analytical solutions for efficient computation.

    • Main Results:

    • The GHSM effectively handles arbitrary-order interactions, expanding the feature space up to the 5th order.
    • Demonstrated the method's effectiveness and efficiency on synthetic data and influenza virus antigenic site identification.
    • Identified meaningful synergistic covariate patterns in influenza virus antigenicity.

    Conclusions:

    • The GHSM is an effective and efficient method for analyzing high-order feature interactions in biological data.
    • The model improves interpretability by uncovering complex synergistic relationships among covariates.
    • The approach has significant implications for understanding biological mechanisms, such as viral antigenicity.