Software

epiqG à¢â‚¬â€œ A flexible software tool for quantitative genetic analysis and prediction in infectious disease

---

Abstract

The concept of genetic selection of individuals that are less susceptible to disease, recover more quickly, and are less infectious has, in recent years, become an active and promising area of research that aims to reduce the spread and impact of infectious diseases in farmed animals. To enable practical genetic selection for these underlying epidemiological host traits from observed disease outbreaks and challenge studies, we have developed computational methods that reliably estimate associated genetic effects. In this presentation we introduce 'qg-epi', a new software tool that can be used to infer genetic parameter estimates (e.g. variance components, breeding values, SNP effects) for the underlying epidemiological traits (susceptibility, infectivity and recoverability) from a wide range of available disease data, and to predict disease transmission in genetically heterogeneous populations. The underpinning epidemiological models can be flexibly tailored to match disease progression and transmission characteristics of the infectious diseases of interest. qg-epi is embedded within the BICI (Bayesian Individual-based Compartmental Inference) platform, which is a general-purpose simulation, inference and posterior simulation tool for individual and population-level compartmental models. qg-epi is therefore a significant upgrade of our previous tool SIRE 2.0, which was restricted to a simple Susceptible-Infectious-Recovered (SIR) disease progression.MODELS: A point-and-click visual interface allows for easy set-up of arbitrary compartmental epidemiological models including flexibility in the nature and number of compartments (e.g. asymptomatic states) and realistic features of infectious disease (e.g. waning immunity, branching, and demographic or spatial stratification). Embedded genetic models accommodate a range of architectures (e.g. polygenic, single genes with large effects) for underlying host traits controlling the transmission dynamics.INFERENCE: qg-epi requires individual-based disease data (e.g. any combination of known transition times, compartmental observations, disease diagnostic test results, or covariates) to infer genetic parameter estimates for the epidemiological traits and makes use of pedigree/genomic relationship matrices. Bayesian posteriors are sampled using an optimised MCMC approach, and outputs include distributions for model parameters, summary statistics, trace plots, correlations, various state visualisations and MCMC diagnostics.IMPLEMENTATION: As well as the visual interface, the user can also interact with the core-code using so-called BICI-script, which allows for batch processing of data. For larger problems, a parallel implementation can be run on multiple cores on a Linux cluster.This presentation explores various different model and data scenarios, shows validation of the methodology and applies qg-epi to a real-world dataset.

Keywords: 2026

How to Cite:

Pooley, C., Marion, G., Prentice, J., Henderson, G. & Doeschl-Wilson, A., (2026) “epiqG à¢â‚¬â€œ A flexible software tool for quantitative genetic analysis and prediction in infectious disease”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2286509. doi: https://doi.org/10.31274/wcgalp.24033

Rights: 1