Figures & data
Figure 1. Full cohort anxiety associations.
(a) Manhattan plot of separation anxiety symptom score associations in full cohort. Dots above dotted line are FDR adjusted p < .05. (b) Scatterplot of GLRX DNAm association with separation anxiety symptom score in full cohort. (c) Scatterplot of ABAT DNAm association with separation anxiety symptom score in full cohort. (d) Scatterplot of XX DNAm association with separation anxiety symptom score in full cohort. (e) Scatterplot of ABAT DNAm association with number of separation anxiety symptoms in full cohort.
Figure 2. Full cohort child had many fears in past 6 months associations.
(a) Manhattan plot of full cohort child had many fears in past 6 months and newborn DNAm. Dots on and above dotted line are FDR adjusted p < .05. (b) Boxplot of newborn DNAm NANS CpG site association with child had many fears in past 6 months (FDR adjusted p = .001). (c) Boxplot of newborn DNAm AMACR CpG site association with child had many fears in past 6 months (FDR adjusted p = .002).
Figure 3. Full cohort child was unhappy/tearful in past 6 months associations.
(a) Manhattan plot of full cohort child was unhappy/tearful in past 6 months and newborn DNAm. Dots on and above dotted line are FDR adjusted p < .05. (b) Boxplot of newborn DNAm NANS CpG site association with child was unhappy/tearful in past 6 months (FDR adjusted p = .0007). The ‘certainly true’ boxplot bottom error bar is in line with the bottom of the interquartile range box. (c) Boxplot of newborn DNAm AMACR CpG site association with child was unhappy/tearful in past 6 months (FDR adjusted p = .002).
Figure 4. Girls only child had many fears in past 6 months associations.
(a) Manhattan plot of targeted CpGs DNAm and child had many fears in the past 6 months in girls. Dots on and above the dotted line are statistically significant at FDR adjusted p < .05. (b) DNAm of NANS CpG site and child had many fears in the past 6 months in girls (FDR adjusted p = .0001). Variable was dichotomized due to small sample size in ‘certainly true’ category in the sex stratified sample. (c) DNAm of HSPA2 CpG site and child had many fears in the past 6 months in girls (FDR adjusted p = .0007).
Figure 5. Girls only child was unhappy/tearful in past 6 months associations.
(a) Manhattan plot of child unhappy/tearful for past 6 months for targeted CpG sites in girls. Dots on and above dotted line are significant at FDR adjusted p < .05. (b) NANS CpG site DNAm and child unhappy/tearful for past 6 months in girls (FDR adjusted p = .001). (c) AMACR CpG site DNAm and child unhappy/tearful for past 6 months in girls (FDR adjusted p = .001). (d) HSPA2 CpG site DNAm and child unhappy/tearful for past 6 months in girls (FDR adjusted p = .023).
Figure 6. Boys’ separation anxiety and child had many fears associations.
(a) Manhattan plot of DNAm and number of separation anxiety symptoms in boys. (b) Scatterplot of LMNB1 CpG site DNAm and number of separation anxiety symptoms in boys. (c) Scatterplot of LMNB1 CpG site DNAm and separation anxiety symptom score in boys. (d) Manhattan plot of DNAm and child had many fears in past 6 months in boys. (e) Boxplot of DAXX CpG site DNAm and child had many fears in past 6 months in boys (FDR adjusted p = .027).
Table 1. Sex interaction results.
Figure 7. Boys’ separation anxiety symptom score mediation Model (LMNB1).
Maternal physical abuse by her father in childhood was the predictor, offspring newborn DNA methylation was the mediator, and male offspring separation anxiety symptom score was the outcome variable. Sobel test indicates complete mediation by DNA methylation of LMNB1 at birth.
Data availability statement
The data that support the findings of this study are available from the Avon Longitudinal Study of Parents and Children (ALSPAC). Restrictions apply to the availability of these data, which were used under licence for this study.