Targeted DNA Methylation Analysis of Embryos

We selected 20 target from the methylKit analysis to validate the WGBS data and analysed these targets in F2 and F3 generation embryos (Supplemental Figure S1). One target (atm) did not show consistent results following standard curve analysis and was discarded. We used 5 replicate pools of embryos per exposure group per generation with this analysis. We used the BisPCR2 method^{59 (link)}, which was adapted at our lab and is extensively described^{16 (link)}, and details can be found in SI materials and methods. Primers were developed using the online Bisearch tool (http://bisearch.enzim.hu/), and validated for specificity and amplicon size by gel electrophoresis, as described previously (Supplemental Table S1)^{16 (link)}. Each primer was validated with standard curve analysis using different ratios of unmethylated DNA to fully methylated DNA. Unmethylated DNA was produced by means of whole genome amplification (Qiagen, Germany) and methylated DNA by M.SssI methyltransferase (New England Biolabs, US)^{16 (link)}. Downstream bioinformatics analysis was performed similarly as with WGBS. Statistical analysis was performed with Seqmonk (v1.36), using the logistic regression filter, with Benjamini-Hochberg FDR correction, with a methylation cut-off of 10%.