Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: Multiclass discovery in array data

Figure 2

The number of discriminatory genes (score) as a function of the cut-off in P value. The data shown is for discovery of two classes in the SRBCT data set. The two curves are for the best partition found (light gray) and for random partitions (dark gray). For the P value cut-off 0.001, the best partition is supported by 602 genes, whereas the expectation for a random partition is 2.3 genes.

Back to article page