Environmental Exposure and DNA Damage

高龄母亲所生的孩子面临各种不良健康结果的风险增加，这表明高龄母亲与孩子DNA的表观遗传变化有关的假设。我们通过对890名挪威新生儿基因组中45万个CpG位点的DNA甲基化测量来探索这一假设，并在KLHL35基因附近发现了一组与母亲年龄显著相关的邻近CpG。我们在一个由1062名挪威新生儿组成的独立小组和200名美国中年妇女中重复了这些发现；这表明这些变化可能在出生后40到60年持续存在。虽然KLHL35基因的功能在很大程度上仍然未知，但这一发现支持了一个假设，即母亲怀孕时的年龄可能会导致她的后代持续出现表观遗传变化，这种变化会持续到成年。

Children born to older mothers are at increased risk of a variety of adverse health outcomes, suggesting the hypothesis that advanced maternal age is associated with epigenetic changes in the childs DNA. We explored this hypothesis using DNA methylation measures at 450,000 CpG sites across the genome in a set of 890 Norwegian newborns and identified a set of adjacent CpGs near the KLHL35 gene that were significantly associated with maternal age. We replicated these finding in an independent group of 1062 Norwegian newborns, and in a set of 200 US middle-aged women; suggesting that these changes may persist 40 to 60 years after birth. While the function of the KLHL35 gene remains largely unknown, this finding supports the hypothesis that a mothers age at pregnancy may lead to persistent epigenetic changes in her offspring that persist into adulthood. Epigenome-wide studies of DNA methylation often make use of high density Illumina methylation arrays. In order to detect small changes associated with exposure or disease, this raw data needs to be processed to remove background noise. We have developed a novel background correction method that uses a mixture of exponential and truncated normal distributions to model signal intensity, and a truncated normal distribution to model background noise. We show that this method is significantly better than other available methods in improving reproducibility and accuracy, resulting in smaller P-value estimates of association for validated CpGs. We incorporate this method, along with a set of additional tools for the analysis of epigenetic data into a software package called ENmix, and make it freely available on the Bioconductor website. Both the existing Illumina 450K array and its successor the MethylationEPIC array use two types of probes to measure DNA methylation across the genome. These two probe types have different chemistries and result in different distributions of methylation values which, if uncorrected, may bias downstream analyses. We developed a novel method using the correlation between adjacent probes to adjust these distributions so that they are more alike, and have incorporated this tool, RCP, into Enmix software.