Skip to main content
Premium Trial:

Request an Annual Quote

This Week in Genome Biology: Dec 26, 2018

Researchers from Johns Hopkins University and elsewhere share a catalog of curated human genes and transcripts. The resource — known as the "Comprehensive Human Expressed Sequences," or CHESS, database — is comprised of 323,258 transcripts representing 20,353 protein coding genes and almost 22,300 non-coding genes gleaned from Genotype-Tissue Expression project experiments, the team reports, along with millions of more tenuous transcripts produced from sites peppered across the genome. "This more comprehensive catalog of genes and splice variants should provide a better foundation for RNA-seq experiments, exome sequencing experiments, genome-wide association studies, and many other studies that rely on human gene annotation as the basis for their analysis," the authors say.

A Stockholm University-led team presents an algorithm called miRTrace, designed for identifying the taxonomic source of organisms analyzed by microRNA sequencing. The two-step approach for linking miRNA sequences back to a clade or species, the researchers explain, which is expected to aid in forensic, food safety, clinical, and research settings. After demonstrating the utility of miRTrace with real and simulated data for more than a dozen plant and animal datasets, for example, they applied the approach to more than 700 publicly available datasets, uncovering and computationally improving contaminated datasets in the process.

Finally, researchers from the Salk Institute for Biological Studies and Molecular Stethoscope report on potential biomarkers for aging, found using a combination of machine learning and RNA sequence data in dermal fibroblast cells from 133 individuals between the ages of 1 year and 94 years. The resulting prediction approach compared favorably with other age prediction methods, they report, including in experiments on 10 individuals with the accelerated aging condition progeria. "Our results suggest that skin fibroblast transcriptome data, coupled with machine learning techniques, can be a useful tool for predicting biological age in humans," the authors conclude. "Applying this approach in a longitudinal study raises the possibility of developing a monitoring and prognostic tool for aging and related diseases."