Picture of neon light reading 'Open'

Discover open research at Strathprints as part of International Open Access Week!

23-29 October 2017 is International Open Access Week. The Strathprints institutional repository is a digital archive of Open Access research outputs, all produced by University of Strathclyde researchers.

Explore recent world leading Open Access research content this Open Access Week from across Strathclyde's many research active faculties: Engineering, Science, Humanities, Arts & Social Sciences and Strathclyde Business School.

Explore all Strathclyde Open Access research outputs...

Local pre-processing for node classification in networks : application in protein-protein interaction

Foley, Christopher Eric and Al Azwari, Sana Mohammad M and Dufton, Mark and Ross, Isla and Wilson, John (2013) Local pre-processing for node classification in networks : application in protein-protein interaction. In: Information Technology in Bio- and Medical Informatics. Lecture Notes in Computer Science . Springer-Verlag Berlin, Heidelberg, pp. 32-46. ISBN 9783642400926

[img]
Preview
PDF
80600032_3_.pdf - Preprint

Download (1MB) | Preview

Abstract

Network modelling provides an increasingly popular conceptualisation in a wide range of domains, including the analysis of protein structure. Typical approaches to analysis model parameter values at nodes within the network. The spherical locality around a node provides a microenvironment that can be used to characterise an area of a network rather than a particular point within it. Microenvironments that centre on the nodes in a protein chain can be used to quantify parameters that are related to protein functionality. They also permit particular patterns of such parameters in node-centred microenvironments to be used to locate sites of particular interest. This paper evaluates an approach to index generation that seeks to rapidly construct microenvironment data. The results show that index generation performs best when the radius of microenvironments matches the granularity of the index. Results are presented to show that such microenvironments improve the utility of protein chain parameters in classifying the structural characteristics of nodes using both support vector machines and neural networks.