Building connections between time series data, outbreak investigation, and phylogenetics to create predictive disease transmission models

Date and time: 2nd Feb 2023, 4-5 pm

Venue: Ramanujan Hall

Host: Siuli Mukhopadhyay

Speaker:  Niket Thakkar

Affiliation: senior research scientist at the Institute for Disease Modeling, within the Bill & Melinda Gates Foundation.

Title: Building connections between time series data, outbreak investigation, and phylogenetics to create predictive disease transmission models

Abstract: 

In this talk, I describe a biologically motivated signal processing approach for building a compartmental, stochastic process model of disease transmission, where the process's mean and variance have distinct dynamics. I apply the approach to COVID-19 time series data from Washington state from January 2020 to March 2021, and I find that the model's hidden states, like population prevalence, agree with survey and other estimates. Then, in the talk's second part, I demonstrate that the same model can be reframed as a branching process with a dynamic degree distribution. This perspective leads to a sampling approach to generate collections of approximate transmission trees, a transmission forest, which I use to estimate some higher order statistics, like the clustering of cases as outbreaks. I find that these predictions are consistent with related observations from outbreak investigations and phylogenetics, suggesting deeper connections between time series volatility and more individualistic measures of disease transmission. Finally, to conclude, I apply similar principles to data on measles transmission in Nigeria, to illustrate the generality of the ideas and to build in this case a more structured model capable of forecasting risk, and I describe how these models are currently being used to inform vaccination strategy.

Commutative Algebra Seminar

Date and time: Thursday 2 February, 2023, 4 pm

Venue: Room 215

Host: Tony Puthenpurakal

Speaker: Tony Puthenpurakal

Affiliation: Mathematics Department, IIT Bombay

Title: A generalization of a theorem of Rees

Abstract:TBA