Table 6. Confirmatory interaction analysis.

The first SNP in an interactive pair				The second SNP in an interactive pair				Combined samples		Independent samples
CHR	RS	NT	GI	CHR	RS	NT	GI	ULR-NOMINAL	ULR-ORDINAL	ULR-NOMINAL	ULR-ORDINAL
								−log10(p)	−log10(p)	−log10(p)	−log10(p)
1	rs618171	A/G	LOC643717	7	rs7805441	C/T	MAGI2	1.2890	1.9355	0.1516	0.0295
6	rs1115620	A/T	LOC643884	12	rs10506451	A/G	PPM1H	0.8380	1.8210	0.0427	0.0225
6	rs2502397	A/G	LOC7805441	12	rs10506451	A/G	PPM1H	0.6946	1.5200	0.1375	0.1086
6	rs1886985	A/G	IMPG1	20	rs6129969	C/T	Intergenic	4.0000	2.3318	1.2248	0.5629
7	rs2214310	C/T	xC7orf9	11	rs7950640	C/G	LOC643652	0.3310	0.7582	0.2480	0.7878
7	rs994531	A/G	xC7orf9	11	rs7950640	C/G	LOC643652	0.4148	0.8589	0.2426	0.6214
7	rs1526555	A/C	KIAA0960	14	rs2331706	C/T	RAD51L1	1.4622	2.1612	0.1350	0.0956
7	rs1526555	A/C	KIAA0960	14	rs765899	C/T	RAD51L1	1.3439	2.0410	0.1430	0.0753

For each SNP pair, chromosome (CHR), RS number (RS), nucleotide types (NT) and gene information (GI) are shown. Based on combined samples or independent samples in CMAS, the interactive effect was examined using a ULR-NOMINAL analysis [model: logit{Prob(Y = YOH|I₁₁, I₁₂,I₂₁,I₂₂,I₁₁×I₂₁,I₁₁×I₂₂,I₁₂×I₂₁,I₁₂×I₂₂,I_Gender,Z_Age)} = a₀+a₁₁×I₁₁+a₁₂×I₁₂+a₂₁×I₂₁+a₂₂×I₂₂+b₁×I₁₁×I₂₁+b₂×I₁₁×I₂₂+b₃×I₁₂×I₂₁+b₄×I₁₂×I₂₂+c×I_Gender+d×Z_Age] or a ULR-ORDINAL analysis [model: logit{Prob(Y = YOH|X₁,X₂,X₁×X₂,I_Gender,Z_Age))} = a₀+a₁×X₁+a2×X₂+a₃×X₁×X₂+c×I_Gender+d×Z_Age], where Y is YOH status, I_ij is an indicator of the three genotypes (j = 1, 2) of SNP i (i = 1, 2), I_Gender is an indicator of gender, Z_Age is a covariate for age, and X_i denotes the number of reference alleles of SNP i (i = 1, 2). The exact p values (in −log₁₀ scale) for the combined and independent samples are shown.