Additional file 2 of A comparison of DNA methylation in newborn blood samples from infants with and without orofacial clefts

Table S1. CpG probes with association P value less than 10−5 in any one of the cleft subtype analysis. Unconditional logistic regression model was used to test for association between DNA methylation M value and cleft status (shared controls vs. each cleft type). Covariables adjusted in each test are control surrogate variables, plate, gender, calendar year of baby’s birth, gestational age, and blood cell subtype composition. Table S2. Sensitivity analysis results by further adjusting for mother’s smoking status, BMI, drinking, folate intake and gestational age. Shown are CpG probes with association P value less than 10−5 in any one of the cleft subtype analysis. Unconditional logistic regression model was used to test for association between DNA methylation M value and cleft status (shared controls vs. each cleft type). Total covariables adjusted in each test are control surrogate variables, plate, gender, calendar year of baby’s birth, gestational age, blood cell subtype composition, smoking, BMI, drinking, folate intake and gestational age. Table S3. CpGs reported to be associated with cleft using Mendelian randomization analysis by Howe et al. 2018. Shown are the analysis results in NCL dataset: mean methylation level in control groups, all cleft case combined group(case), and case subgroups; case-control logistic regression estimated coefficient and P values for the combined case group and case subgroups. Table S4. Methylation means and association P values (controls compared to combined cleft cases) in Alvizi et al. (Scientific Reports, 2016) study and NCL study. Shown are the overlapped CpGs between NCL study and 578 CpGs reported by Alvizi et al. Table S5. Candidate gene results. Shown are CpGs around literature reported cleft-associated genes with methylation association P value less than 0.01 in any one of the cleft subtype analysis. Table S6. Differentially methylated regions (DMR) with Sidak multiple testing corrected P value