DrIgor Siveroni

Development of software tools for the modelling and analysis of infectious disease dynamics, and for the analysis and interpretation of genetic sequence data from important pathogens such as HIV, Ebola virus and Influenza.

I am currently part of a team working on the development of an individual-based model simulator of TB written in C. The simulator models the transmission dynamics of TB and provides a framework for the introduction and assessment of health interventions.

Previously, I developed PhyDyn, a BEAST 2.5 (Bayesian Evolutionary Analysis by Sampling Trees) module/plug-in that implements coalescent models for mulit-deme populations with nonlinear dynamics, based on the theory and methods initially presented in Complex population dynamics and the coalescent under neutrality (Volz EM, 2012, Genetics, Vol:190), and extended in Bayesian Phylodynamic Inference with Complex Models (Volz EM and Siveroni I, 2018, PLOS Computational Biology).

PhyDyn: Epidemiological modelling in BEAST.
PhyDyn is a BEAST2 package for performing Bayesian phylogenetic inference under models that deal with structured populations with complex population dynamics. This package enables simultaneous estimation of epidemiological parameters and pathogen phylogenies.

PhyDyn implements a structured coalescent model for a large class of epidemic processes specified by a deterministic nonlinear dynamical system, and computes the log-likelihood of a gene genealogy conditional on a complex demographic history. Genealogies are specified as timed phylogenetic trees in which lineages are associated with the distinct subpopulation in which they are sampled. Epidemic models are defined by a series of ordinary differential equations (ODEs) specifying the rates that new lineages introduced in the population (birth matrix) and the rates at which migrations, or transition between states occur (migration matrix).

The package's underlying coalescent model (where birth and migration rates are a function of time and the underlying population dynamics) and rich ODE syntax (polynomials, timed conditional expressions, trigonometric functions) enables the specification and implementation of a large class of epidemiological processes. The framework can be applied to models with spatial structure, multiple stages of infections and models of vector-borne diseases and other multi-host pathogens.

Documentation and source available here: https://github.com/mrc-ide/PhyDyn.

Poster presentation at the MIDAS Annual Networking meeting (2017).