. Author manuscript; available in PMC: 2012 May 1.

Published in final edited form as: Stat Med. 2010 Feb 20;29(4):489–503. doi: 10.1002/sim.3815

Table 3.

The means and standard errors of the posterior probabilities of being binding target genes ( $\hat{\Pr} (I_{i x} = 1)$ ), probabilities of being differentially expressed genes for non-binding target genes ( $\hat{\Pr} (DE ∣ I_{i x} = 0)$ ), probabilities of being differentially expressed genes for binding target genes ( $\hat{\Pr} (DE ∣ I_{i x} = 1)$ ), probabilities of being potential genes for non-binding target genes ( $\hat{\Pr} (I_{i z} = 1 ∣ I_{i x} = 0)$ ), probabilities of being potential genes for binding target genes ( $\hat{\Pr} (I_{i z} = 1 ∣ I_{i x} = 1)$ ) from 10 simulated data sets for each set-up. The prior distribution for p_x is Beta(200, 800).

Simulations	$\hat{\Pr} (I_{i x} = 1)$	$\hat{\Pr} (DE ∣ I_{i x} = 0)$	$\hat{\Pr} (DE ∣ I_{i x} = 1)$	$\hat{\Pr} (I_{i z} = 1 ∣ I_{i x} = 0)$	$\hat{\Pr} (I_{i z} = 1 ∣ I_{i x} = 1)$
Setup 1	0.214±0.003	0.004±0.001	0.865±0.021	0.152±0.031	0.966±0.006
Setup 2	0.198±0.005	0.032±0.007	0.878±0.035	0.639±0.051	0.640±0.061
Setup 3	0.204±0.006	0.201±0.012	0.173±0.050	0.365±0.085	0.929±0.032
Setup 4	0.218±0.003	0.005±.001	0.833±0.031	0.169±0.023	0.955±0.007