|
Activity Number:
|
328
- Advances in MCMC Theory and Practice
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
IMS
|
|
Abstract #316711
|
|
|
Title:
|
Scalable Bayesian Inference for Phylodynamics
|
|
Author(s):
|
Julia A. Palacios* and Sifan A. Liu
|
|
Companies:
|
Stanford University and Stanford University
|
|
Keywords:
|
scalable ;
Bayesian;
phylodynamics;
coalescent;
nonparametrics
|
|
Abstract:
|
Bayesian phylodynamics aims to infer the evolutionary dynamics of populations from observed molecular sequence variation in a sample of individuals of a population. However current implementations are not scalable for large sample sizes. One key modeling aspect in phylodynamics is that data dependency is modeled through a latent genealogical tree. While most scalable Bayesian strategies are developed for iid data, we show that divide-and-conquer strategies can be successfully applied for salable Bayesian inference in phylodynamics. Finally, we show the applicability of our approach to estimating the evolutionary dynamics of SARS-CoV-2.
|
Authors who are presenting talks have a * after their name.
