. Author manuscript; available in PMC: 2014 Jul 31.

Published in final edited form as: Schizophr Res. 2011 Jul 14;131(0):43–51. doi: 10.1016/j.schres.2011.06.023

Table 4.

Test of independent effects between rs2274736/rs22401756 and rs7147796^a.

Markers	Haplotype	Frequency	OR (Alternative)	OR (Null)	Sub-Null P	Null P
rs2274736–rs7147796	CC	0.029	ref	ref	0.2660	0.0244
	TC	0.454	1.11	ref
	CG	0.312	0.96	0.91	0.0188
	TG	0.206	1.06	0.91
rs2401756–rs7147796	AC	0.029	ref	ref	0.3290	0.0227
	GC	0.454	1.10	ref
	AG	0.311	0.95	0.91	0.0143
	GG	0.206	1.05	0.91

In these conditional analyses, the null hypothesis is that the haplotypes having the same alleles at the conditioned locus, the second locus, rs7147796, have same effect on disease risk. The alternative hypothesis is that all haplotypes may have different disease risks. The sub-null hypothesis tests whether the two haplotypes sharing the conditioned allele have the same disease risks. Significant P values are shown in bold.