An agent-based model to study the effects of epidemiological factors on three common Influenza strains’ virulence

Keywords

Agent-based modeling; Influenza; epidemiology

Abstract

Influenza is a virus that has caused harm at a devastating level worldwide. The biotechnology industry spends billions of dollars researching and developing new vaccines for Influenza every year due to the abrupt antigenic shifts and the subtle antigenic drifts. Currently, there are a variety of Influenza vaccines that are prescribed based on one’s age and health. The goal of this project is to identify the influence of discrete factors, such as vaccination rates, and temperature on the overall rate of different Influenza strains’ infection in the South Atlantic region for a given year. This was accomplished by developing an agent-based model to simulate the progression of the virus from the 2010 through 2016 seasons. Although similar models have been completed before on Influenza, the objective of this project was to analyze the differences and similarities of three common strains of Influenza based upon their contagiousness. The three common strains analyzed throughout this study were H1N1, H3, and H3N2. The biological differences between these three strains are found in the variations in the binding molecules, hemagglutinin and neuraminidase. To study such variations, an agent-based model was created using the NetLogo software to simulate the spread of the different Influenza strains amongst individuals in the South Atlantic region. For the purpose of this project, 3 agent-based models, one for each strain, of the same functionality were developed. The agent based model developed accounts for the behavior and effectiveness of the actual annual Influenza vaccine. That is, it accounts for one to create an immune response to the most common Influenza strains predicted for the upcoming season and that one only maintains protection for various strains if the specific sequence is accounted for in the vaccine cocktail. Alternatively, individuals can also create immunity to a specific strain by getting infected with the specific strain and creating immunity for it. The contagiousness, R0, of each strain for a particular time was calculated by . The vaccination rate, V0, of Influenza per season was calculated by . In running the model, the agents can be infected by the specific strain, have a set immunity via being vaccinated, or become immune to the specific strain by getting infected and recovering from the infection. In the model, the infected individuals are marked as a red agent, the healthy individuals are marked as a white agent, and the individuals who have recovered from a specific strain of Influenza are marked as a blue agent. The results obtained from the proposed model demonstrated the behavioral similarities and differences of the three Influenza strains amongst the South Atlantic region population. Such understanding will help doctors and researchers better understand the effectiveness of the Influenza vaccine as well as the virulence of the different subtypes of Influenza. Therefore, this simulation could help medical professionals visualize, analyze, and understand if there is a need to reposition the Influenza vaccine in the future.

Date Created

January 2017

This document is currently not available here.

Share

COinS