Table 1.

Analysis of general, unconstrained variance vs. nested, constrained variance models

Data	Model*	Genes	s / gene**	Mm	Parameters	BIC***
Townsend et al. (2003)	AU	4506	5.2	33312.6	7	-11147.1
Townsend et al. (2003)	AC	5759	4.1	21296.8	4	-4108.8
Townsend et al. (2003)	AV	5759	4.8	21365.8	4	-4039.8
Townsend et al. (2003)	MU	4506	7.0	33969.2	7	-10490.6
Townsend et al. (2003)	MC	5759	5.7	21469.2	4	-3936.4
Townsend et al. (2003)	MV	5759	6.2	21459.1	4	-3946.5
Sudarsanam et al. (2000)	AU	4756	3.5	19874.9	5	-1053.6
Sudarsanam et al. (2000)	AC	5888	3.2	16199.2	3	502.9
Sudarsanam et al. (2000)	AV	5888	3.6	16200.4	3	504.1
Sudarsanam et al. (2000)	MU	4756	4.4	20229.8	5	-698.7
Sudarsanam et al. (2000)	MC	5888	4.1	16494.4	3	798.0
Sudarsanam et al. (2000)	MV	5888	4.7	16370.7	3	674.3

* (A)dditive or (M)ultiplicative error, with (U)nconstrained variances, a common (C)oefficient of variation, or a common (V)ariance. ** seconds of processor time on a dual 1 GHz PowerPC G4 *** Bayesian Information Criterion