An international team led by investigators at the University of Exeter outline apparent links between altered biological aging and mutations affecting epigenetic regulators, particularly DNA methyltransferase or histone methyltransferase enzymes. The researchers did array-based methylation profiling in related and unrelated individuals with Tatton-Brown-Rahman syndrome, who carried known alterations in the DNA methyltransferase-coding gene DNMT3A. Along with enhanced DNA methylation-based aging in these individuals, their results revealed lower-than-usual methylation at development, differentiation, morphogenesis, and malignancy-related genes. The authors subsequently saw additional epigenetic aging shifts in individuals with NSD1 or KMT2D histone methyltransferase gene alterations. "Together," they report, "our findings provide fundamentally new insights into aberrant epigenetic mechanisms, the role of epigenetic machinery maintenance, and determinants of biological aging in these growth disorders."
A Pennsylvania State University team presents a machine-leaning method called CN-Learn for identifying copy number variants in exome sequence data with dialed-down false-positive calls. By learning from CNV calls from several available detection algorithms, the machine-learning method picks out high-confidence CNVs, the researchers say. When they trained CN-Learn on four existing CNV calling algorithms and applied it to exome sequences for 503 "Simons Variation in Individual Project" participants, for example, they found that the approach narrowed in on verified CNVs with roughly 90 percent precision and 85 percent recall. "Our results suggest that a small set of high-quality validated CNVs and an objective machine learning method can help alleviate several shortcomings of existing integration approaches to generate an informed set of clinically relevant CNVs," the authors write.
Finally, researchers from Ohio State University and the University of Iowa describe a long-read sequencing-based strategy for mapping long-range chromatin accessibility and nucleosome dynamics. The method — known as "methyltransferase treatment followed by single-molecule long-read sequencing," or MeSMLR-seq — "offers direct measurements of both nucleosome-occupied and nucleosome-evicted regions on a single DNA molecule, which is challenging for many existing methods," they write. The team used MeSMLR-seq and RNA sequencing to map nucleosome organization and chromatin heterogeneity in individual haploid yeast cells, ultimately phasing chromatin profiles in and around several genes, quantifying chromatin accessibility, and uncovering ties to the expression of specific genes.