Table 2

Independent lead SNPs that reached genome-wide significance under four different measures of early-phase insulin response

Chr	SNP	Phenotype	Position	Location	Gene	Effect/noneffect allele	Effect allele frequency^a	Imputation quality^a	Discovery cohort			Replication cohort
Chr	SNP	Phenotype	Position	Location	Gene	Effect/noneffect allele	Effect allele frequency^a	Imputation quality^a	Effect size per allele (%)^b	SE	P	Effect size per allele (%)^b	SE	P
3	rs115404340	insulin₃₀	67839179	Intron	SUCLG2-AS1	G/A	0.01	0.44	78.5	0.098	3.8E-9	−15.7	0.13	0.19
9	rs505922	ΔInsulin	136149229	Intron	ABO	C/T	0.32	1	−6.5	0.012	3.3E-8	−7.8	0.021	7.2E-5^*
9	rs657152	IGI	136139265	Intron	ABO	A/C	0.34	0.98	8.1	0.011	3.0E-12	2.7	0.017	0.11
19	rs74889068	IGI	46199363	Intron	QPCTL	A/G	0.15	0.93	−8.8	0.016	3.8E-9	−2.1	0.024	0.36
3	rs115404340	IR_BMIadj	67839179	Intron	SUCLG2-AS1	G/A	0.01	0.42	72.5	0.092	3.4E-9	−9.8	0.12	0.41
19	rs74889068	IR_BMIadj	46199363	Intron	QPCTL	A/G	0.15	0.93	−10.9	0.018	4.4E-11	−4.0	0.028	0.14

Threshold for genome-wide significance is 5 × 10⁻⁸. Chr, chromosome.
* A successful replication, with P value < 0.05.
a In the discovery cohort.
b β-Coefficient expressed as the percentage difference in the outcome by one copy of the effect allele.