Classification in biological networks with hypergraphlet kernels

Abstract

Biological and cellular systems are often modeled as graphs in which vertices represent objects of interest (genes, proteins, drugs) and edges represent relational ties between these objects (binds-to, interacts-with, regulates). This approach has been highly successful owing to the theory, methodology and software that support analysis and learning on graphs. Graphs, however, suffer from information loss when modeling physical systems due to their inability to accurately represent multi-object relationships. Hypergraphs, a generalization of graphs, provide a framework to mitigate information loss and unify disparate graph-based methodologies. We present a hypergraph-based approach for modeling biological systems and formulate vertex classification, edge classification and link prediction problems on (hyper)graphs as instances of vertex classification on (extended, dual) hypergraphs. We then introduce a novel kernel method on vertex- and edge-labeled (colored) hypergraphs for analysis and learning. The method is based on exact and inexact (via hypergraph edit distances) enumeration of hypergraphlets; i.e., small hypergraphs rooted at a vertex of interest. We empirically evaluate this method on fifteen biological networks and show its potential use in a positive-unlabeled setting to estimate the interactome sizes in various species. Availability https://github.com/jlugomar/hypergraphlet-kernels/ Supplementary information Supplementary data are available at Bioinformatics online.

(Hyper)graph Kernels over Simplicial Complexes

“…To the best of our knowledge, the only attempt to bridge the gap between hypergraphs and kernel methods is described in [61], where the authors proposed edit distance-based hypergraphlet kernels for node and vertex classification and link prediction, e.g., in protein-protein interaction networks.…”

Beyond the dyad: uncovering higher-order structure within cohesive animal groups

“…Adopting a higher-order approach allows to effectively represent interactions and or associations that occur in groups of different sizes, helping in better understanding the structural differences between interacting agents that belong to different domains. For example, higher-order motif analysis (Lugo-Martinez et al . 2021; Lotito et al .…”

“…Adopting a higher-order approach allows to effectively represent interactions and or associations that occur in groups of different sizes, helping in better understanding the structural differences between interacting agents that belong to different domains. For example, higher-order motif analysis (Lugo-Martinez et al 2021;Lotito et al 2022) revealed that clustered pairwise collaborations among scientific authors are predictive of future 3-way interactions (Benson et al 2018), and that protein or chemical compounds that interact in groups typically cannot do so in single pairs with other agents. In general, the decomposition of large hyperlinks (e.g.…”