The causal effect of red blood cell folate on genome-wide methylation in cord blood: a Mendelian randomization approach

Table 1 Genes in proximity to loci with a significant change in methylation levels per one unit change in log (RBC folate)

Symbol	Chr	UCSC annotated features	CpG island	Δ β-value	Bootstrap 95% CI
OBFC2B	12	Promoter associated	Shore	0.386	(0.227, 0.551)
GPR75; LOC100302652	2	5′UTR; 1st Exon; Enhancer	Island	0.349	(0.184, 0.515)
AIRE	21	Within 100 bp of TSS	Island	0.267	(0.094, 0.444)
MIMT1; PEG3;ZIM2	19	Within 500 bp of TSS; 5′UTR; 5′UTR	Island	0.207	(0.088, 0.332)
CIRBP; C19orf24	19	Body; Within 1000 bp of TSS	Island	0.176	(0.097, 0.256)
SMG6; SRR	17	Within 500 bp of TSS; 5′UTR	Shore	0.125	(0.039, 0.214)
GPR12	13	5′UTR	Island	-0.095	(-0.161, -0.030)

For each significant locus, including RefSeq symbol for gene(s) in closest proximity, chromosome, UCSC annotated feature, relation to UCSC CpG island. Effect estimates were reported for sites for which ${\hat{β}}_{IV}$ from the two-stage least squares (TSLS) modeling the square-root arcsine beta-values were significant at the 0.05 α-level, adjusting for whether conception was planned. Without this variance stabilizing transformation, the effect estimates are unbiased, but inference will not be correct. For interpretability, effect estimates were calculated among the significant sites without the transformation and 95% confidence interval for 1000 bootstrapped replicates were reported.

ISSN: 1471-2105