VERA: agent-based modeling transmission of antibiotic resistance between human pathogens and gut microbiota

• Published in: Bioinformatics Vol. 35; no. 19; pp. 3803 - 3811
• Main Authors: Glushchenko, Oksana E, Prianichnikov, Nikita A, Olekhnovich, Evgenii I, Manolov, Alexander I, Tyakht, Alexander V, Starikova, Elizaveta V, Odintsova, Vera E, Kostryukova, Elena S, Ilina, Elena I
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 01-10-2019
• ISSN: 1367-4803; 1460-2059; 1367-4811
• DOI: 10.1093/bioinformatics/btz154

Abstract
Motivation
The resistance of bacterial pathogens to antibiotics is one of the most important issues of modern health care. The human microbiota can accumulate resistance determinants and transfer them to pathogenic microbiota by means of horizontal gene transfer. Thus, it is important to develop methods of prediction and monitoring of antibiotics resistance in human populations.
Results
We present the agent-based VERA model, which allows simulation of the spread of pathogens, including the possible horizontal transfer of resistance determinants from a commensal microbiota community. The model considers the opportunity of residents to stay in the town or in a medical institution, have incorrect self-treatment, treatment with several antibiotics types and transfer and accumulation of resistance determinants from commensal microorganism to a pathogen. In this model, we have also created an assessment of optimum observation frequency of infection spread among the population. Investigating model behavior, we show a number of non-linear dependencies, including the exponential nature of the dependence of the total number of those infected on the average resistance of a pathogen. As the model infection, we chose infection with Shigella spp., though it could be applied to a wide range of other pathogens.
Availability and implementation
Source code and binaries VERA and VERA.viewer are freely available for download at github.com/lpenguin/microbiota-resistome. The code is written in Java, JavaScript and R for Linux platform.
Supplementary information
Supplementary data are available at Bioinformatics online.