. 2010 Aug 16;5(8):e12183. doi: 10.1371/journal.pone.0012183

Table 4. Statistics estimates and P values for major haplotypes showing significant associations at the 0.05 level with at least one smoking-related phenotype under the additive and dominant models in the male sample.

SNP Number	SNP combinations	Haplo-type	Freq	SI			SQ			SC
				Esti-mate	P	Global P	Esti-mate	P	Global P	Esti-mate	P	Global P
18-19-20	rs16969948-rs481134-rs951266	TCC	0.77	−0.102^a	0.034^a	0.023^a	−0.108^a	0.069^a	0.003^a	−0.050^a	0.401^a	0.276^a
19-20-21	rs481134-rs951266-rs514743	CCT	0.82	−0.173^a	0.020^a	0.037^a	−0.186^a	0.005^a	0.001^a	−0.011^a	0.869^a	0.529^a
20-21-22	rs951266-rs514743-rs6495308	CTG	0.74	−0.114^a−0.265^d	0.015^a 0.025^d	0.008^a 0.137^d	−0.124^a−0.307^d	0.032^a 0.031^d	0.003^a 0.158^d	−0.077^a	0.186^a	0.303^a
21-22-23	rs514743-rs6495308-rs950776	TGA	0.74	−0.118^a	0.011^a	0.071^a	−0.134^a	0.019^a	0.060^a	−0.082^a	0.155^a	0.411^a
22-23-24	rs6495308-rs950776-rs11072768	GAA	0.70	−0.132^a	0.003^a	0.038^a	−0.140^a	0.011^a	0.103^a	−0.107^a	0.054^a	0.288^a
23-24-25	rs950776-rs11072768-rs7166158	AAT	0.57	−0.101^a	0.018^a	0.024^a	−0.116^a	0.029^a	0.112^a	−0.072^a	0.180^a	0.161^a
		ACT	0.12	0.168^a0.238^d	0.009^a 0.001^d	0.024^a 0.013^d	0.173^a0.214^d	0.026^a 0.014^d	0.112^a 0.117^d	0.162^a0.210^d	0.043^a 0.019^d	0.161^a 0.154^d
24-25-26	rs11072768-rs7166158-rs8043123	ATC	0.38	−0.094^a−0.141^d	0.032^a 0.029^d	0.010^a 0.073^d	−0.047^a	0.393^a	0.208^a	−0.050^a	0.359^a	0.009^a
		CTC	0.15	0.157^a0.149^d	0.010^a 0.035^d	0.010^a 0.073^d	0.159^a	0.031^a	0.208^a	0.159^a	0.036^a
26-27-29	rs8043123-rs4887077-rs11638372	CGG	0.53	−0.229^d	0.022^d	0.035^d	−0.134^d	0.128^d	0.021^d	0.122^d	0.158^d	0.369^d

(1) SI = smoking initiation; SQ = smoking quantity; SC = smoking cessation; (2) Significant associations at the 0.05 level before Bonferroni correction for multiple testing are given in bold and those after Bonferroni correction for multiple testing are given in bold and underlined (corrected P value at a 0.05 significance level is 0.0125 under the assumption of a maximum of four major haplotypes for each SNP combination); (3) Superscripts indicate the genetic model used: a = additive; d = dominant; and (4) For each sample, age, and area were used as covariates.