In Genome Biology this week, a trio of researchers reports their application of a human functional protein interaction network to cancer data analysis. Specifically, "by extending curated pathways with non-curated sources of information including protein-protein interactions, gene coexpression, protein domain interaction, gene ontology annotations and text mined protein interactions," the team constructed a network covering nearly 50 percent of the human proteome. They then applied their network to the investigation of two genome-wide glioblastoma multiforme — and several other cancer — data sets.
Stanford University investigators report their method for "modeling non-uniformity in short-read rates in RNA-seq data," in Genome Biology. The authors suggest two models to predict constant rates along each transcript based on local sequences. "These models explain more than 50 percent of the variations and can lead to improved estimates of gene and isoform expressions for both Illumina and Applied Biosystems data," the authors write.
A large, international research team describes their computational re-analysis of existing whole-genome sequence data using their novel microarray-based approach. "Our results indicate that a large number of SVs [structural variants] have been unreported in the individual genomes published to date," the authors write. The team detected 12,178 structural variants — covering 40.6 megabases — which "were not reported in the initial sequencing of the first published personal genome," they authors write.
Investigators at the Duke University School of Medicine and their colleagues compare exome, whole-genome, and whole-transcriptome sequencing. They write that while WGS is "the most complete," it is still the most expensive; RNA-seq, they suggest, "is a fast and inexpensive alternative approach for finding coding variants in genes with sufficiently high expression levels," though "a high false positive rate can be problematic when working with RNA-seq data, especially at higher levels of coverage."