Table 3.

Functional enrichment in correctly predicted positive genes. Predefined motifs: the models built with predefined motifs. Auto motifs: the models built with auto motifs. Up: the models built for up-regulated genes. Down: the models built for down-regulated genes. #Gp: the number of Gp genes (positive genes that are predicted as positive by a model). #Gn: the number of Gn genes (positive genes that are predicted as negative by a model). : the number of enriched functional categories in the set of Gp genes. : the number of enriched functional categories in the set of Gn genes. : the number of enriched functional categories in the set of Gp' genes (see text). The sample sizes for up- and down-regulated genes in cell cycle are 56 and 59, respectively, and 167 and 164 in stress conditions. The ranges represent individual 95% confidence interval for the means.

Model			#Gp	#Gn
Predefined motifs	Cell cycle	Up	17.2 ± 2.2	27.1 ± 2.7	15.1 ± 2.9	8.2 ± 2.6	12.3 ± 3.5
		Down	14.0 ± 2.1	31.0 ± 2.5	14.5 ± 3.7	9.6 ± 3.4	8.4 ± 3.3
	Stress	Up	16.4 ± 1.0	33.1 ± 1.1	13.4 ± 2.3	10.6 ± 1.8	10.3 ± 1.9
		Down	21.0 ± 2.0	30.7 ± 1.7	17.1 ± 2.4	7.8 ± 1.7	14.7 ± 2.3
Auto motifs	Cell cycle	Up	24.9 ± 1.8	19.3 ± 1.8	16.9 ± 3.2	7.9 ± 2.5	13.9 ± 3.2
		Down	23.6 ± 1.9	21.4 ± 1.8	15.6 ± 3.7	8.4 ± 2.9	11.8 ± 3.5
	Stress	Up	27.0 ± 0.8	22.6 ± 0.8	16.2 ± 2.3	9.1 ± 1.9	15.2 ± 2.4
		Down	29.4 ± 1.3	22.2 ± 1.1	18.2 ± 2.3	6.6 ± 1.5	16.4 ± 2.4