Clustering microarray data using the Self Organising Oscillator Network

Clustering algorithms belong to an area of research that has many practical uses. Over the years, many different clustering algorithms have been proposed. Of these, the majority that are in common use today tend to be based on mathematical techniques which utilise the density of the data in data space. This has advantages for many scenarios, however there are occasions where density based clustering algorithms may not always be the most appropriate choice. The Self-Organising Oscillator Network (SOON) is a comparatively new clustering algorithm [1], that has received relatively little attention so far. The SOON is distance based, meaning that clustering behaviour is different in a number of ways that can be beneficial. This paper examines the performance of the SOON with a biological dataset taken from microarray experiments on the Cell-cycle of yeast. The SOON is shown to be a useful addition to the available clustering algorithms, being able to highlight small (but potentially significant) clusters of interest in a dataset.