GO term over-representation analysis of k-means clustered genes.

Gene Ontology terms enrichment analysis was carried out comparing the GO term frequency distribution into each cluster against that in the whole microarray set (hypergeometric statistics, p<0.05). Only the lowest node per branch of the hierarchical structure of the Gene Ontology that fulfills the filter condition - cut off 2 sequences- was reported. Cluster 1 and 4 were merged into a unique group as they presented the same temporal expression patterns and only differed for the intensity. Shown are: Cluster, the number of cluster obtained from k-means analysis (see Figure 3); Level, level in the GO tee of biological processes; GO Term, over-represented feature; N, number of mussel sequences associated to each GO term; Gene ID, EMBL accession number of each sequence found.