Additional file 1 of DNA methylation mediates the effect of maternal smoking on offspring birthweight: a birth cohort study of multi-ethnic US mother–newborn pairs

Additional file 1: Table 1. Genome-wide DNA methylation association study identified 38 CpG sites significantly associated with maternal smoking during pregnancy in 954 mother–newborn pairs from the Boston Birth Cohort. Additional file 1: Table 2. Genome-wide DNA methylation association study with maternal smoking during pregnancy: Comparison of significant CpGs among total sample, Blacks only, and non-Blacks only. Additional file 1: Table 3. Methylation mediation effect of each of the 38 single CpG sites on the maternal smoking-newborn birthweight association in 954 mother–newborn pairs from the Boston Birth Cohort. Additional file 1: Table 4. Single CpG, single gene score, and combined multiple gene mediation analysis on the maternal smoking—newborn birthweight association, among 504 male newborns in the Boston Birth Cohort. Additional file 1: Table 5. Single CpG, single gene score, and combined multiple gene mediation analysis on the maternal smoking—newborn birthweight association, among 450 female newborns in the Boston Birth Cohort. Additional file 1: Table 6. Single CpG, single gene score, and combined multiple gene mediation analysis on the maternal smoking—newborn birthweight association, among 679 Black newborns in the Boston Birth Cohort. Additional file 1: Table 7. Single CpG, single gene score, and combined multiple gene mediation analysis on the maternal smoking—newborn birthweight association, among 275 non-Black newborns in the Boston Birth Cohort. Additional file 1: Table 8. Characteristics of mother–newborn pairs included (N = 954) versus excluded (N = 7555) from this current study. Additional file 1: Figure 1. Illustration of the relationship between maternal smoking categories (0 = never, 1 = quitter, 2 = current) and GFI1 gene score (Panel A), AHRR cg05575921 (Panel B), and CYP1A1 gene score (Panel C); and between smoking categories (0 = never, 1 = quitter, 2 = current) and birthweight (Panel D), in 954 mother–newborn pairs from the Boston Birth Cohort. Additional file 1: Figure 2. Manhattan and Q-Q plots for the EWAS analyses in cord blood in associations with maternal smoking, in Black (Fig 2A) and non-Black subset (Figure 2B). Grey dotted line represents the epigenome-wide significance cut-off after Bonferroni correction, and the red dotted line represents the epigenome-wide significance cut-off with FDR< 0.05; Genes in red represent known genes associated with smoking as identified in previous studies. Additional file 1: Figure 3. An illustration of our sequential analytical methods to dissect the interplay of maternal smoking (X), cord blood DNA methylation (M: mediator), and newborn birthweight (Y), where C’ represents the direct effect of X on Y. Additional file 1: Figure 4. Distribution of cotinine/hydroxy-cotinine by smoking status: Panel A: cord plasma cotinine/hydroxy-cotinine by maternal smoking status; Panel B: maternal plasma cotinine/hydroxy-cotinine by maternal smoking status. Additional file 1: Figure 5. Flowchart of Study Participants.