Activity Number:
|
606
- Genetic Data for Epidemiologic Inference During an Outbreak: Statistical Challenges and Solutions
|
Type:
|
Invited
|
Date/Time:
|
Thursday, August 2, 2018 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistics in Epidemiology
|
Abstract #326753
|
|
Title:
|
Estimation and Comparison of Transmission Trees Using Sequence Data
|
Author(s):
|
Michelle Kendall and Caroline Colijn*
|
Companies:
|
Oxford University and Simon Fraser University
|
Keywords:
|
transmission;
phylogenetics;
metrics;
Bayesian
|
Abstract:
|
Estimating who infected whom with the help of sequence data presents many challenges. In-host diversity, variable times and durations of infectiousness, the presence of unsampled individuals and the possibility that more than one pathogen lineage is transmitted between hosts all complicate the relationship between who infected whom and the genealogical relationships among pathogen sequences. We describe several approaches to this problem, including an approach that handles unsampled cases and within-host diversity. We describe the effects of layering additional types of data on to the likelihood of each transmission tree - in this way we can incorporate data on the timing of symptoms and on case locations. There are now a number of Bayesian inference methods available to reconstruct who infected whom using sequence data; these make different assumptions and require different input data. We present a metric on transmission trees which allows direct comparison of posterior transmission trees reconstructed with different methods. It also allows us to find a median tree for each posterior collection or cluster of trees.
|
Authors who are presenting talks have a * after their name.