Principal component analysis and hierarchical clustering of Arabidopsis transcriptome data. (A) Principal component analysis is an exploratory technique used to describe the structure of high dimensional data, e.g. derived from microarrays, by reducing its dimensionality. Here, expression values for 22.800 genes in 8 tissue/cell types are projected onto the first three principal components. The first principal component separates pollen and root hairs from the other tissues, while the second and third principal components show a further, though less significant, separation of the samples. (B) Hierarchical clustering is used to group similar objects into “clusters”, producing a tree (called dendrogram) that shows the hierarchy of the clusters. The dendrogram shows a clear separation of a pollen and root hair cluster from a cluster including the other sample types.