On analytical approaches to epidemics on networks

Area of Science:

• Network science
• Epidemiology
• Graph theory

Background:

• Traditional random graph models lack control over triangle formation, which is prevalent in real-world networks like social networks.
• Triangles in networks significantly influence the spread of infections and disease dynamics.

Purpose of the Study:

• To develop a method for constructing random graphs with specified degree distributions and a controlled number of triangles.
• To analyze the spread of two distinct types of infections on these precisely constructed random graphs.
• To establish bounds for key epidemiological measures like R(0) and extinction probability.

Main Methods:

• Construction of random graphs with a given degree distribution and a specified expected number of triangles.
• Analysis of infection spread using two model infections: fixed infectious period and all-or-none transmission.
• Application of these models to derive theoretical bounds for epidemiological parameters.

Main Results:

• Successfully generated random graphs with controllable triangle counts, overcoming limitations of standard models.
• Demonstrated the utility of these graphs in analyzing infection dynamics under different transmission scenarios.
• Provided a framework for deriving upper and lower bounds for R(0) and extinction probabilities.

Conclusions:

• The developed random graph model offers a more realistic representation of networks for studying epidemic spread.
• This method provides a robust analytical tool for understanding disease transmission and evaluating control strategies.
• The findings contribute to a deeper understanding of network-influenced epidemiological dynamics.