A trio from the University of Toronto presents a new approach for analyzing alternative polyadenylation (APA) from RNA sequencing data in Genome Biology. The method, called QAPA for quantification of APA, relies on estimates of alternative 3′ UTR expression and a resource of annotated poly(A) sites that can uncover UTR sequences affected by APA. They report that their approach is faster and more sensitive than others, and that when they applied it to a neuronal differentiation dataset, they were able to uncover and characterize APA in a range of conditions.
Pittsburgh researchers, meanwhile, have developed an algorithm called SQUID to predict transcriptomic structural variations (TSVs) from RNA-seq alignments and report that it is typically about 20 percent more accurate than WGS-based SV detection methods. When they applied the algorithm to 401 tumor samples from The Cancer Genome Atlas, the researchers found that breast invasive carcinoma had the largest variance in TSVs/non-fusion-gene TSVs per sample and uncovered novel non-fusion-gene TSVs that affect known tumor suppressor genes.
Also in Genome Biology, researchers from the Mayo Clinic report on their characterization of glioma stem cells. In particular, the researchers assessed the epigenome of glioma stem cells isolated from glioblastoma patient-derived xenografts. They found that abnormal expression of TET family members was linked to global levels of 5mC and 5fC/5caC. Additionally, they report that glioma and neural stem cells respond differently to TET expression, which is reflected by differences in their epigenomes.