Table 1. Signals associated with levels of quantitative glycemic traits among non-diabetic subjects in CHNS.

Locus	SNP	Chr	Position	EA/NEA	EAF	Imputation r2	Genotype	Beta	SE	P-value
Fasting Glucose
SIX3-SIX2	rs895636	2	45,188,353	T/C	0.40	GT	2124/2709/953	0.099	0.018	2.3 x 10−8
G6PC2	rs34177044	2	169,754,485	A/G	0.37	0.70	2043/2935/808	0.145	0.021	6.9 x 10−12
G6PC2	rs2232326	2	169,764,491	T/C	0.96	GT	5281/488/17	0.252	0.042	1.8 x 10−9
HbA1c
FN3KRP	rs9895455	17	80,683,980	T/C	0.49	0.97	1821/3504/1853	0.095	0.018	1.0 x 10−7
Fasting Insulin
CNTN6	rs13078376	3	932,902	C/T	0.69	0.69	2637/2688/484	0.124	0.024	3.2 x 10−7

Results are shown if the association signal reached genome-wide significance (P <5x10^-8) or is the strongest association if no signals reached genome-wide significance. Trait values were adjusted for age, age², BMI, PC1, and sex. Chromosome positions are based on hg19. Effect alleles are associated with higher trait values. Beta estimates reflect per allele effects of variants on inverse normal transformed traits. CHNS, China Health and Nutrition Survey; chr, chromosome; EA, effect allele; NEA, noneffect allele; EAF, effect allele frequency; SE, standard error; GT, genotyped