MATria: a unified centrality algorithm

Trevor Cickovski,Vanessa Aguiar-Pulido,Giri Narasimhan

doi:10.1186/s12859-019-2820-7

Trevor Cickovski, Vanessa Aguiar-Pulido + Show 1 more

Open Access

https://doi.org/10.1186/s12859-019-2820-7

Copy DOI

Abstract

BackgroundComputing centrality is a foundational concept in social networking that involves finding the most “central” or important nodes. In some biological networks defining importance is difficult, which then creates challenges in finding an appropriate centrality algorithm.ResultsWe instead generalize the results of any k centrality algorithms through our iterative algorithm MATRIA, producing a single ranked and unified set of central nodes. Through tests on three biological networks, we demonstrate evident and balanced correlations with the results of these k algorithms. We also improve its speed through GPU parallelism.ConclusionsOur results show iteration to be a powerful technique that can eliminate spatial bias among central nodes, increasing the level of agreement between algorithms with various importance definitions. GPU parallelism improves speed and makes iteration a tractable problem for larger networks.

Highlights

Computing centrality is a foundational concept in social networking that involves finding the most “central” or important nodes
We checked some of the agreed important genes discovered by MATRIA in network B
MATRIA found genes for Death-Associated Protein 3 (DAP3) which has been marked essential in other eukaryotic organisms for its critical roles in respiration and apoptosis [20], and the Heat Shock Protein (HSP) which has been marked essential for apoptosis in both prokaryotes and eukaryotes [21] and is involved in protein folding [22]

Summary

Introduction

Computing centrality is a foundational concept in social networking that involves finding the most “central” or important nodes. In some biological networks defining importance is difficult, which creates challenges in finding an appropriate centrality algorithm. The concept of centrality is fundamental to social network theory and involves finding the most important or central nodes in a social network. Betweenness centrality [1] bases importance on the number of shortest paths over all pairs of nodes that run through a node (finding hubs in a network), closeness [2] on the overall length of the shortest paths towards all other nodes that start from a node (finding nodes in the “center” of a network), and degree [3] on the number of connections. Google’s PageRank [5] models centrality by a random walker which probabilistically either moves to a neighbor or someplace

Results

Discussion

Conclusion