Intensity-based hierarchical Bayes method improves testing for differentially expressed genes in microarray experiments

Table 1 Simulated estimation of prior degrees of freedom for SMT and IBMT

Dependency strength of variance on intensity	Method	d₀ = 1	d₀ = 4	d₀ = 16	d₀ = 100
Low	SMT	0.200	0.494	0.774	0.923
	IBMT	0.200	0.500	0.800	0.963
Middle	SMT	0.198	0.472	0.703	0.813
	IBMT	0.200	0.501	0.800	0.961
High	SMT	0.194	0.422	0.571	0.630
	IBMT	0.200	0.500	0.801	0.962

Values listed are for the function d₀/(d₀ + d_g) and are the mean of 100 simulations. Perfect values for each prior degrees of freedom are: d₀ = 1: 0.20, d₀ = 4: 0.50, d₀ = 16: 0.80, and d₀ = 100: 0.962

ISSN: 1471-2105