An R package for alignment-free network comparison
Speakers: Martin O’Reilly (The Alan Turing Institute, UK); Gesine Reinert (University of Oxford and The Alan Turing Institute, UK)
Date: 24 August 2017
Venue: The Alan Turing Institute
The representation of data as a network has become common across many disciplines, with network representations proving powerful in areas from finance to biology to cyber security. The ability to compare networks of different sizes and densities has therefore become an increasingly important and challenging area of research. There are only a small number of tools that have been developed for this type of comparison, two of which originate from research led by Turing Fellows Gesine Reinert and Charlotte Deane.
These network comparison methods, called Netdis and NetEMD, are not based on network alignment but rather on the subgraph content of the networks. They have sparked interest in several other scientific groups and companies as they conceptually are able to compare very large networks very quickly.
Having developed initial proof of concept code implementing Netdis and NetEMD, the researchers were keen to make their new comparison methods accessible to a wider range of users. The Turing Research Engineering team worked with them to make the Netdis and NetEMD methods available as an open, robust and easy to use software package for R, one of the most widely used programming languages for data analysis. This will enable a wide range of groups across and outside academia to easily use these cutting edge network comparison methods in their own work.
Gesine Reinert (Oxford University and The Alan Turing Institute, UK)
Gesine Reinert is a University Lecturer at the Department of Statistics, Oxford, and Fellow at Keble College, Oxford (2000 – present). Since 2016 she is the Vive-chair of the EU-funded COST Action COSTNET: the European Cooperation for Statistics of Network Data Science.
Her current and main research interests are in network statistics and to investigate such networks in a statistically rigorous fashion. Often this will require some approximation, and approximations in statistics are another of her research interests.
Gesine is Turing Fellow and a Fellow of the IMS.
Martin O’Reilly (The Alan Turing Institute, UK)
Martin O’Reilly is Principal Research Software Engineer at the Turing. Martin is part of the Turing’s Research Engineering team, a group of data scientists and software engineers who work with Turing researchers to increase the impact of their work by applying it to real world problems and turning research solutions into software that can be easily used by others.