. 2020 Aug 5;10(8):138. doi: 10.3390/life10080138

Table A2.

Summary of performance for each method at various cutoff points when the total sample size = 1,000,000 and $r r ~ U (1.2, 10)$ .

True Signal Ratio		0.03			0.05			0.1
Method	Cutoff *	Power	Sensitivity	PPV	Power	Sensitivity	PPV	Power	Sensitivity	PPV
ROR	1	0.997	0.853	0.391	0.996	0.829	0.569	1.000	0.779	0.810
	1.5	0.997	0.800	0.649	0.996	0.772	0.755	1.000	0.711	0.886
	2	0.997	0.745	0.763	0.996	0.710	0.834	1.000	0.645	0.916
	2.5	0.997	0.691	0.818	0.996	0.645	0.871	1.000	0.575	0.932
	3	0.997	0.640	0.852	0.996	0.587	0.892	1.000	0.502	0.941
PRR	1	0.997	0.853	0.391	0.996	0.829	0.569	1.000	0.779	0.810
	1.5	0.997	0.800	0.650	0.996	0.772	0.755	1.000	0.711	0.886
	2	0.997	0.745	0.763	0.996	0.709	0.835	1.000	0.644	0.916
	2.5	0.997	0.691	0.819	0.996	0.645	0.871	1.000	0.574	0.932
	3	0.997	0.638	0.852	0.996	0.585	0.893	1.000	0.501	0.941
IC	$\log_{2} (1)$	0.994	0.837	0.263	0.992	0.817	0.451	1.000	0.777	0.794
	$\log_{2} (1.5)$	0.992	0.766	0.748	0.990	0.739	0.859	1.000	0.686	0.960
	$\log_{2} (2)$	0.989	0.694	0.912	0.988	0.663	0.954	1.000	0.600	0.983
	$\log_{2} (2.5)$	0.986	0.651	0.943	0.988	0.622	0.967	0.998	0.550	0.987
	$\log_{2} (3)$	0.983	0.556	0.971	0.987	0.525	0.983	0.994	0.438	0.991
LRT	0.2	0.969	0.664	0.982	0.982	0.642	0.992	0.988	0.571	0.997
	0.1	0.966	0.645	0.992	0.978	0.623	0.996	0.986	0.550	0.999
	0.05	0.959	0.625	0.996	0.971	0.605	1.000	0.982	0.532	0.999
	0.025	0.953	0.607	0.998	0.971	0.589	1.000	0.977	0.513	0.999
	0.01	0.949	0.584	1.000	0.967	0.567	1.000	0.972	0.492	1.000
GPS	1	0.954	0.641	0.998	0.967	0.621	0.999	0.980	0.590	0.999
	1.5	0.954	0.639	0.998	0.967	0.619	0.999	0.979	0.583	0.999
	2	0.953	0.625	0.998	0.967	0.587	0.999	0.978	0.524	1.000
	2.5	0.953	0.577	0.998	0.965	0.522	1.000	0.974	0.422	1.000
	3	0.952	0.501	0.998	0.964	0.428	1.000	0.963	0.314	1.000
sB	1	0.976	0.756	0.753	0.983	0.743	0.892	0.990	0.683	0.977
	1.5	0.958	0.640	0.991	0.968	0.617	0.998	0.976	0.545	1.000
	2	0.941	0.537	1.000	0.962	0.515	1.000	0.970	0.426	1.000
	2.5	0.921	0.450	1.000	0.949	0.420	1.000	0.958	0.324	1.000
	3	0.889	0.372	1.000	0.938	0.334	1.000	0.939	0.237	1.000
BCPNN	$\log_{2} (1)$	0.970	0.716	0.825	0.971	0.681	0.937	0.968	0.609	0.989
	$\log_{2} (1.5)$	0.961	0.635	0.994	0.966	0.583	0.997	0.961	0.508	1.000
	$\log_{2} (2)$	0.953	0.545	1.000	0.962	0.491	0.999	0.955	0.411	1.000
	$\log_{2} (2.5)$	0.941	0.472	1.000	0.955	0.409	1.000	0.943	0.326	1.000
	$\log_{2} (3)$	0.926	0.406	1.000	0.948	0.335	1.000	0.932	0.247	1.000
TreeScan	0.2	0.972	0.688	0.980	0.979	0.651	0.992	0.985	0.574	0.999
	0.1	0.966	0.673	0.988	0.977	0.632	0.996	0.983	0.553	0.999
	0.05	0.964	0.658	0.992	0.974	0.615	0.997	0.983	0.533	1.000
	0.025	0.959	0.646	0.996	0.966	0.603	1.000	0.977	0.518	1.000
	0.01	0.955	0.627	0.997	0.963	0.581	1.000	0.968	0.500	1.000

ROR, Reporting Odds Ratio; PRR, Proportional Reporting Ratio; IC, Information Component; LRT, Likelihood ratio test; GPS, Gamma Poisson Shrinker; BCPNN, Bayesian Confidence Propagation Neural Network; sB simplified Bayes; TreeScan, Tree-based Scan Statistic; * Cutoff values for the lower bound of the 95% CI for ROR, PRR, IC, BCPNN, and sB, for EB05 for GPS, and for the p-value for LRT and TreeScan.