Skip to main content
Premium Trial:

Request an Annual Quote

Genome Research Studies on AML Methylation Types, Natural Antisense Transcripts, STARR-Seq

Researchers in the US and Germany describe more than a dozen "epitypes" with distinct DNA methylation features in acute myeloid leukemia (AML). Using array-based profiling of methylation marks across the genome in samples from 649 individuals with AML, the team tracked down 13 clusters with comparable methylation features, including epitypes with pronounced or diminished methylation and those sharing ties to specific tumor mutations. Clusters with stem cell-like methylation features and stem cell-like expression activity tended to have lower-than-usual survival compared to cases marked by other epitypes, the authors report, while the shift from one epitype to another at relapse appeared to be rare. "These results demonstrate that DNA methylation-based classification integrates important molecular features of AML," they write, "to reveal the diverse pathogenic and biological aspects of the disease."

An international team led by the University of Liverpool takes a look at antisense RNA contributions to vertebrate development, focusing on natural antisense transcripts (NATs) for long non-coding RNAs (lncRNAs) in developing zebrafish. Using RNA-seq, CAGE-seq, and computational biology, the investigators defined two groups of co-expressed NATs involved in early development, dubbed group-1 and group-2 NATs. "In contrast to group-1, which is enriched in genes involved in developmental pathways, group-2 protein-coding genes are enriched in house-keeping functions," they report, noting that additional features suggest that group-1 NATs "function post-transcriptionally to silence spurious expression of developmental genes."

Finally, investigators at Duke University outline a statistical modeling strategy for refining the search for regulatory elements from "self-transcribing active regulatory region sequencing" (STARR-seq) data by dialing down technical signal biases. The approach, called "correcting reads and analysis of differentially active elements" (CRADLE), relies on generalized linear regression to account for biases, the authors explain, before picking up significant regulatory activity. "This approach substantially improves precision and recall over current methods, improves detection of both activating and repressive regulatory elements, and controls for false discoveries despite strong local correlations in signal," they suggest, noting that CRADLE "substantially improves the use of STARR-seq by providing a robust estimation of regulatory activity and improved visualization of raw signals."