ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

APPLYING HYPERGRAPHS TO STUDIES IN QUANTITATIVE BIOLOGY

Journal: LIFE: International Journal of Health and Life-Sciences (Vol.9, No. 2)

Publication Date:

Authors : ;

Page : 21-33

Keywords : Hypergraph; Hypergraph Model; Hypergraph Classifier; Graph;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

The objective of this research is to demonstrate hypergraph versatility and applicability for modeling diverse biological systems. The inherent structure of hypergraphs allows for encoding of higher-order feature interactions, providing a flexible framework for efficient models that can enhance our understanding of physical phenomena and one that can be generalized across various datasets. By adopting innovative methods including centrality measure and populations of models rather than singular instances, biases and overfitting tendencies are mitigated, again presenting promise for application across a broad spectrum of biological systems. Furthermore, emphasis is placed on the significance of probabilistic distribution analysis in elucidating threshold selection and feature relevance while maintaining high levels of accuracy. Our results demonstrate the advantages of hypergraph models on two different datasets; with the first on gene expression and the identification of outlier genes and the second on classifying starch grains. There is significant scope in the application of the hypergraph to a wider class of biological systems, with the potential to improve understanding of the biological processes.

Last modified: 2024-06-27 15:30:58