Abstract
This work presents the results of applying two clustering techniques to gene expression data from the mussel Mytilus galloprovincialis. The objective of the study presented in this paper was to cluster the different genes involved in the experiment, in order to find those most closely related based on their expression patterns. A self-organising map (SOM) and the k-means algorithm were used, partitioning the input data into nine clusters. The resulting clusters were then analysed using Gene Ontology (GO) data, obtaining results that suggest that SOM clusters could be more homogeneous than those obtained by the k-means technique.