Extra colloquium Mathematics: Mirko Signorelli, MSc

Title: Inferring Community-driven Structure in Complex Networks

Abstract:

What have our genes in common with deputies in a Parliament, or with the employees of a research institute? They constantly interact with their peers, creating complex networks.

The study of networks and relational data, which can be dated back until the Thirties, has received a sudden boost at the turn of the century. Nowadays, network science is a subject of interest in a wide range of scientific fields, including sociology, biology, computer science, engineering and – last but not least - statistics.

In this talk I will consider three different problems that arise in cell biology, in political science and in epidemiology, and I will discuss three statistical models that we have designed to address them.

I will start by presenting NEAT, a network-based hypergeometric test that allows to integrate traditional gene enrichment analysis tests with information on known interactions between genes.

Then, I will discuss how some extensions to stochastic block models can be combined with a penalized inference approach to reconstruct collaborations between parties from data on bill cosponsorships in the Italian Parliament.

I will conclude by presenting a novel clustering method for populations of networks. I will discuss two different strategies for its estimation, as well as its application to data on daily contact networks collected in a French research institute.

Colloquium coordinators are Prof.dr. A.J. van der Schaft ( a.j.van.der.schaft rug.nl ),

Dr. A.V. Kiselev (e-mail: a.v.kiselev rug.nl )

http://www.rug.nl/research/jbi/news/colloquia/mathematics-colloquia/