Table 2.

Difference of Synonymous Substitution Rate (K₄) Between Housekeeping and Tissue-Specific Genes

	Human—mouse									Mouse—rat
Breadth measure	Housekeeping^a	Tissue-specific^a	ΔK₄^b	P^c	N	Housekeeping^a	Tissue-specific^a	ΔK₄^b	P^c	N
K₄
EST	0.387	0.502	0.115	0.00077	121	0.161	0.18	0.019	0.00067	463
SAGE	0.411	0.475	0.064	0.0026	394	0.157	0.175	0.018	<0.00001	1928
Microarray	0.387	0.501	0.114	0.00005	103	0.165	0.172	0.007	0.057	789
K₄ residuals from regression on K_A
EST	0.024	0.055	0.047	121	-0.031	0.0099	0.0012	0.41	463	0.0087
SAGE	-0.002	0	0.51	394	-0.002	0.0092	0.0065	0.032	1928	0.0027
Microarray	0.026	0.048	0.05	103	-0.022	0.0067	-0.0012	0.61	789	0.0079

↵a Average K₄ or average residuals of K₄. Residuals were calculated from expected K₄ values, which were predicted from linear regression of log(K₄) on log(K_A) including all genes. Genes were classified as housekeeping/tissue-specific if supported by experiments in both human and mouse for the human—mouse comparison, and by experiments in mouse for the mouse—rat comparison. Only non-disparate gene pairs were included.
↵b Difference in K₄ (or residuals) between tissue-specific and housekeeping averages.
↵c Probability of finding an equal or greater difference in 100,000 randomized genomes.

Genomic Regionality in Rates of Evolution Is Not Explained by Clustering of Genes of Comparable Expression Profile