Skip to main content

Table 4 Sensitivity analysis results for the two-study and five-study simulation data. Sensitivity analysis results for the prior degrees of freedom assigned to the slide and experiment effect variance parameters, φ j g 2 MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciaacaGaaeqabaqabeGadaaakeaaiiGacqWFgpGzdaqhaaWcbaGaemOAaOMaem4zaCgabaGaeGOmaidaaaaa@323F@ and σ j g 2 MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciaacaGaaeqabaqabeGadaaakeaaiiGacqWFdpWCdaqhaaWcbaGaemOAaOMaem4zaCgabaGaeGOmaidaaaaa@3249@ , respectively, for the two-study and five-study simulation data sets with medium mean τ g 2 MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciaacaGaaeqabaqabeGadaaakeaaiiGacqWFepaDdaqhaaWcbaGaem4zaCgabaGaeGOmaidaaaaa@30EE@ = 0.3 (differentially expressed); 0.03 (non-differentially expressed) and simulated percent differentially expressed genes p s = 10%. Shown are true integration-driven discovery rate (t IDR) and true integration-driven revision rate (t IRR) for threshold value of posterior probability of differential expression γ = 0.95, and the number of true discovered genes for posterior expected false discovery rate pe FDR = 5%, for the Bayesian standardized expression integration (SEI) and probability integration (PI) models.

From: Bayesian meta-analysis models for microarray data: a comparative study

Two-Study Simulation Data True Integration-Driven Discovery Rate, γ= 0.95 True Integration-Driven Revision Rate, γ= 0.95 True Genes, pe FDR = 5%
Prior Degrees of Freedom       
φ j g 2 MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciaacaGaaeqabaqabeGadaaakeaaiiGacqWFgpGzdaqhaaWcbaGaemOAaOMaem4zaCgabaGaeGOmaidaaaaa@323F@ σ j g 2 MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciaacaGaaeqabaqabeGadaaakeaaiiGacqWFdpWCdaqhaaWcbaGaemOAaOMaem4zaCgabaGaeGOmaidaaaaa@3249@ SEI PI SEI PI SEI PI
3 3 10.8% 16.2% 15.9% 4.5% 185 211
6 6 4.4% 11.9% 11.6% 1.2% 197 216
10 10 3.6% 10.2% 7.9% 0.6% 202 218
20 20 1.7% 8.1% 4.4% 0.5% 206 219
40 40 1.6% 6.6% 5.9% 0.5% 208 221
Five-Study Simulation Data       
3 3 2.2% 15.2% 10.2% 3.6% 212 241
6 6 0% 8.8% 13.0% 2.8% 219 246
10 10 0% 6.9% 13.1% 2.7% 225 248
20 20 0% 3.9% 14.4% 3.1% 228 250
40 40 0% 5.1% 12.8% 1.8% 227 250