An Australian and American team takes a look at sites showing distinct cytosine methylation patterns from one parental allele to the next. The researchers relied on long-read nanopore sequence data to search for and characterize imprinted DNA methylation sites in mouse embryonic placenta samples based on methylation levels in the context of haplotype information — an allele-specific methylation detection approach they went on to validate with short read reduced-representation bisulfite sequencing and with methylation data reported in the past. "Our analysis successfully identifies known imprinting control regions, as well as some novel differentially methylated regions which, due to their proximity to hitherto unknown monoallelically expressed genes, may represent new [imprinting control regions]," they report.
Researchers from the Mayo Clinic's Center for Individualized Medicine consider the origins of germline structural variants and their interactions with chromatin organization and epigenomic features in the human genome. Starting with almost 27,000 structural variants identified from the 1000 Genomes Project, the team mapped the distributions of structural variants arising through distinct processes, ranging from non-allelic homologous recombination (NAHR) to deletions or insertions by transposable elements or retrotransposons. When they analyzed these structural variants alongside genome features such as chromatin accessibility, the authors saw "biased distribution of [structural variants] across genome compartments, epigenetic states, and functional annotations," which appear to stem from a "complex interplay of differential mechanistic biases, chromosome conformation, DNA accessibility, and natural selection."
Finally, a team from Germany attempts to tally authentic human microRNAs using small RNA sequence data, in silico analyses, and additional experimental validation. The researchers began by mapping hundreds of billions of small RNA-seq reads from nearly 29,000 datasets to the human genome, identifying more than 24,100 candidate mature miRNAs. A significant proportion of those appeared to be false-positive, they note, but follow-up blotting analyses in human cell lines provided a closer look at dozens of known and 119 new mature miRNA candidates produced from 108 miRNA precursors. "In total, we extrapolated 2,300 true human mature miRNAs, 1,115 of which are currently annotated in miRbase V22," they write, noting that additional experimental validation in the future will help revise miRNA targets modeled with false-positive miRNAs.