Pleiotropic association of LIPC variants with lipid and urinary 8-hydroxy deoxyguanosine levels in a Taiwanese population

Table 5 Mediation tests of the TG levels on the associations between the LIPC genotypes and HDL-C levels in male participants

		rs2043085	rs1800588	rs1532085
Criterion 1	α regression coefficient Standard error P value
Criterion 1	α regression coefficient Standard error P value	0.059 0.021 0.005	0.063 0.021 0.004	0.061 0.021 0.005
Criterion 2	β regression coefficient Standard error ^#P value
	β regression coefficient Standard error ^#P value	−0.181 0.021 7.2 × 10⁻¹⁶	−0.183 0.021 2.9 × 10⁻¹⁶	− 0.181 0.021 6.5 × 10^− 16
	γ’ regression coefficient Standard error *P value
	γ’ regression coefficient Standard error *P value	0.022 0.008 0.005	0.026 0.008 0.001	0.022 0.008 0.004
Criterion 3	αβ + γ’ regression coefficient Standard error P value
Criterion 3	αβ + γ’ regression coefficient Standard error P value	0.011 0.008 0.194	0.014 0.009 0.100	0.011 0.009 0.183
Criterion 4	αβ regression coefficient Standard error P value (Sobel test)
Criterion 4		−0.011 0.004 0.008	−0.012 0.004 0.005	− 0.011 0.004 0.006

LIPC genotypes were analyzed in dominant models,
α: unstandardized coefficient for the association between LIPC genotypes and TG levels
β: unstandardized coefficient for the association between TG and HDL-C levels (when adjusting for LIPC genotypes)
Direct effect = γ’, Total effect = αβ + γ’, Mediation (indirect) effect = αβ
P: adjusted for age, sex, BMI, current smoke; *: indicated the P value of association between SNPs and HDL-C after adjusted for age, sex, BMI, current smoke and TG; ^#: indicated the P value of association between TG and HDL-C after adjusted for age, sex, BMI, current smoke and SNPs