0000000001265203
AUTHOR
Rocío Urdinguio
MOESM12 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 12: Figure S4. Boxplots showing the DNA methylation beta values of the 36 common 0→5→10 dmCpGs described in Table 2.
MOESM11 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 11: Figure S3. Treemap plots indicating the results of REViGO sematic analyses of significantly enriched (FDR < 0.05) gene ontology biological process terms for genes containing (a) hyper- and (b) hypomethylated 0→5 dmCpGs. The dmCpGs are grouped by annotated genomic location (see Table S1, column “annotation”, for the annotations; “promoter” is formed by collapsing “Distal promoter” and “Promoter (<= 1kb)”, “gene body” is formed by collapsing “3’ UTR”, “5’ UTR”, “Intron”, “Exon” and “Downstream”). See Table S6 for full results, including Molecular Function and Cellular Component terms.
MOESM10 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 10: Figure S2. a) Treemap plots indicating the results of REViGO sematic analyses of significantly enriched (FDR < 0.05) gene ontology biological process terms for genes that simultaneously contained 0→5 hyper- and hypomethylated dmCpGs. In total, 460 significant terms were found (see Table S6 for full results, including Molecular Function and Cellular Component terms, and also 5→10 dmCpG enrichments). b) Equivalent plots for genes containing, respectively, hyper- or hypomethylated dmCpGs, irrespective of those same genes also containing dmCpGs that changed in the opposite direction.
MOESM5 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 5: Figure S1. Boxplots indicating the distribution of absolute beta values of the DNA methylation changes for 0→5 and 5→10 hyper- and hypomethylated dmCpGs. Effect size is measured as median difference and Cliff’s delta.
MOESM13 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 13: Table S7. DNA methylation values obtained by bisulfite sequencing for 3 CpGs across the original 33 subjects (technical validation) and two independent longitudinal cohorts (biological validation).
MOESM4 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 4: Table S6. Gene ontology enrichment analysis of 0→5 and 5→10 dmCpGs. Enrichments were calculated from the difference between the dmCpGs obtained in each of the analyses and the GO ontology database by the R/Bioconductor package missMethyl. Ontologies for genes with 1) both hyper- and hypomethylated probes 2) exclusively either hyper- or hypomethyated and 3) from mapped hyper- or hypomethylated dmCpGs which did not take into account CpGs differentially methylated in the opposite direction in the same genes are included. Also included are the ontologies found when analyzing 0→5 dmCpGs grouped by gene region (promoter, exon, intron and gene body). All of the analyses include …
MOESM7 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 7: Table S3. Histone mark enrichment analysis of 0→5 and 5→10 dmCpGs. Enrichments were calculated based on differences between the dmCpGs obtained in each of the analyses and the full collection of Roadmap epigenomics hg19 regions integrated in the LOLA extended software. Corresponding array backgrounds were used for the different comparisons.
MOESM9 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 9: Table S5. Enhancer element enrichment analysis of 0→5 and 5→10 dmCpGs. Enrichments were calculated from the difference between the dmCpGs obtained in each of the analyses and the enhancer elements obtained from the EnhancerAtlas database. A customized LOLA database which included information related to the enhancers in the different tissues/cell lines and corresponding array backgrounds were used for the appropriate enrichment calculation.
MOESM6 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 6: Table S2. Blood cell-type compositions as predicted by the Houseman algorithm for the 33 samples.
MOESM3 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 3: Table S1. Lists of annotated 0→5 and 5→10 dmCpGs.
MOESM2 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 2: Table S8. Primer sequences for bisulfite sequencing of the validated CpGs.
MOESM8 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 8: Table S4. Chromatin state enrichment analysis of 0→5 and 5→10 dmCpGs. Enrichments were calculated based on differences between the dmCpGs obtained in each of the analyses and the hg19 chromatin segmentation regions (ChromHMM, 18 states) obtained from Roadmap and ENCODE consortia. A custom LOLA database including information related to the chromatin states in the different tissues/cell lines and corresponding array backgrounds were used for the correct enrichment calculation.
MOESM1 of Longitudinal genome-wide DNA methylation analysis uncovers persistent early-life DNA methylation changes
Additional file 1: Table S9. Clinical characteristics of the 18 subjects from the independent cohorts.