Figure 4From: Combining transcriptional datasets using the generalized singular value decompositionThe average number of successfully identified microarray genes, using distance within three arraylets as the measure of similarity (light bars). On the right hand axis the calculated p-values characterizing the expected number of false positives for a single GSVD are shown. The dark bars indicate the result obtained if those microarray genes not used in the GSVD are randomized.Back to article page