Random networks, epidemics and phylogenetics


Project leader


Funding source

Swedish Research Council - Vetenskapsrådet (VR)


Project Details

Start date: 01/01/2007
End date: 31/12/2009
Funding: 1950000 SEK


Description

The project covers to areas of applied probability and associated statistical inference: 1. Mathematical modelling of infectious disease spread. One aim is to derive asymptotic results concerning the risk for eventual outbreaks and its size, for more realistically structured communities using theory for random networks. In particular, we will adress the concepts of degree (the social activity is known to vary greatly in human populations) and clustering (complete social subgraphs are more common in reality than occur by pure chance, e.g. households). Here an important question is also to determine effective vaccination policies corresponding to thinning the graph in various ways. Other topics to be studied are: stochastic models allowing for varying infectious response, models for contact tracing and their inference to be applied to tuberculosis data, and inference procedures for real-time analysis of new infections. 2. Phylogenetics - evolutionary relationships between organisms. One project here is to improve the mathematical argument for the empirically observed phenomenon that bootstrap support values are lower than the corresponding Bayesian support for estimated phylogenetic trees. Another project is to derive estimation techniques for ultrametric (real time) trees relaxing the assumption of a molecular clock. This means that we must allow for variation in average substitution rate for different branches (species) and over time in the underlying evolutionary tree.

Last updated on 2017-24-03 at 12:09