Motifs in triadic random graphs based on Steiner triple systems.

Area of Science:

• Network science
• Graph theory
• Computational biology

Background:

• Complex networks are conventionally viewed as built from pairwise links.
• Subnetwork patterns, or motifs, are increasingly recognized as fundamental network building blocks.
• Existing generative models lack the capability to test the functional roles of subgraph motifs.

Purpose of the Study:

• To develop sound generative models for complex networks based on triadic substructures.
• To address the challenge of independent specification of triad patterns in network models.
• To enable the investigation of the functional implications of motif statistics.

Main Methods:

• Utilizing exponential random graph models (ERGMs) framework.
• Employing Steiner triple systems (STSs) to define independent triad specifications.
• Combining ERGMs and STSs to create generative models for network ensembles.

Main Results:

• Generated networks with non-trivial triadic Z-score profiles.
• Identified inherent statistical correlations between triad pattern abundances.
• Analytically calculated degree distributions for the generated triadic random graphs.

Conclusions:

• The developed models provide a new tool for studying the functional significance of network motifs.
• Understanding statistical correlations in motif abundance is crucial for interpreting network properties.
• The findings advance the field of network science by offering a method to generate and analyze networks based on subgraph structures.