Table 5.

Revised predictions based on recent results from the Psychiatric Genomics Consortium with the prior effect size distribution estimated for cancer risk loci

SNP	Disorder	Cases	Controls	One-sided P-value	Prediction intervals for $P_{rep}$
					Conjugate Bayesa	Mixture Bayesa	Mixture Bayesb	Lazerroni et al.
rs2535629	ADHD	2787	2635	0.1005	(0.023, 0.977)	(0.023, 0.977)	(0.023, 0.977)	(2.57e-5, 0.93)
	ASD	4949	5314	0.098	(0.022, 0.977)	(0.022, 0.977)	(0.022, 0.977)	(2.39e-5, 0.93)
	BP	6990	4820	3.305e-06	(0.017, 0.977)	(0.017, 0.977)	(5.89e-9, 0.93)	(1.69e-13, 0.04)
	MDD	9227	7383	0.000108	(0.016, 0.977)	(0.016, 0.977)	(7.41e-5, 0.98)	(4.89e-11, 0.18)
	Schizophrenia	9379	7736	3.355e-05	(0.015, 0.977)	(0.015, 0.977)	(9.33e-7, 0.95)	(6.92e-12, 0.11)
	All	33,332	27,888	1.27e-12	(0.001, 0.871)	(0.001, 0.871)	(8.91e-20, 4.2e-5)	(7.41e-23, 1.17e-5)

ADHD attention deficit-hyperactivity disorder, ASD autism spectrum disorder, BP bipolar disorder, MDD major depressive disorder

^aThe prior effect size distribution using the conjugate model with the variance estimated based on the tabulated values of effect sizes reported in Park et al.

^bThe prior effect size distribution specified directly by the estimates reported in Park et al.