Table 4. Enrichment by protein function of ANOVA-identified genes.

Protein class	Actual	R	Expected	Ratio	p-value	z-score	In data set	In protein function	Protein function in database
Transcription factors	45	959	17.13	2.63	4.02E-09	6.934	10.56%	4.69%	4.02%
Phosphatases	7	230	4.109	1.70	1.20E-01	1.446	1.64%	3.04%	0.96%
Ligands	13	514	9.183	1.42	1.34E-01	1.285	3.05%	2.53%	2.16%
Kinases	16	654	11.68	1.37	1.29E-01	1.292	3.76%	2.45%	2.74%
Enzymes	54	2724	48.67	1.11	2.26E-01	0.8195	12.68%	1.98%	11.42%
Proteases	11	559	9.987	1.10	4.15E-01	0.3272	2.58%	1.97%	2.34%
Receptors	30	1565	27.96	1.07	3.71E-01	0.4026	7.04%	1.92%	6.56%
Other	253	16691	298.2	0.84	1.68E-06	-4.822	59.39%	1.52%	70.00%

Over representation analysis of the 403 ANOVA interaction-identified genes was performed based on broadly defined protein functions. The Metacore database for the human genome annotates 23,844 network objects for the background list. Column abbreviations: Actual, number of ANOVA identified objects for a given class; R, number of background objects for a given class; Expected, mean value for hypergeometric distribution (403*R/23,844); Ratio, the ratio of actual/expected; p-value, probability to have the given actual value or higher (or lower for negative z-score); In data set, percent of objects of a given function in the ANOVA dataset; In protein function, percent of ANOVA objects of a given function relative to the number of background objects with the same function; Protein function in database, percentage of ANOVA objects of a given function in the background list.