Modelling Preferential and Incomplete Sampling in Genomic Surveillance
- Discipline
- Genomic epidemiology; infectious disease modelling; statistical methodology
- Methodology
- Stochastic modelling; Bayesian computation; simulation experiments; model comparison; analysis of pathogen genomic data
- Date
- 30 Jul 2026
- Description
- Genomic surveillance datasets rarely represent a random sample of infections: sequencing effort changes over time, locations and population groups, and may increase during recognised outbreaks. This project will extend birth–death–sampling models to represent realistic sampling mechanisms and assess how incomplete or preferential sampling affects estimates of transmission dynamics. The work will use simulated and empirical datasets to compare alternative observation models and determine when explicitly modelling the sampling process materially improves inference.
- Keywords
- Genomic surveillance;preferential sampling;phylodynamics;observation models
- Would Suit Applicants Who
- Have quantitative training and an interest in Bayesian statistics, infectious disease epidemiology, population genetics or scientific computing
