A Belgian-led group introduces an online, literature-mining method called Beegle that's designed to detect previously overlooked gene-disease associations in an automated manner. The researchers' newly described search and discovery method begins with a literature-mining approach that links genes to a particular query. Coupled with the learning method Endeavour, which analyzes genomic data to prioritize and rank genes to produces new hypotheses. In their own hands, the Beegle pipeline picked up 69 percent of the top gene associations in the Online Mendelian Inheritance in Man and 84 percent of the top 100 OMIM genes.
Researchers from Germany, the UK, and Spain describe a database for human embryonic stem cell and induced pluripotent stem cell lines. The public registry, dubbed hPSCreg, includes background information, legal considerations, ethical standards, and biological data on almost 700 human embryonic stem cell lines and 76 induced pluripotent stem cell lines generated in more than two dozen countries. "hPSCreg is the first global registry that holds both, manually validated scientific and ethical information on hPSC lines," the team writes, "and provides access by means of a user-friendly, mobile-ready web application."
Finally, a team from the NorthShore University HealthSystem and the University of Chicago presents a statistical modeling method for gleaning somatic mosaicism based on paired-end sequence data from tumor samples and samples from healthy, unaffected individuals. The ultra-fast computational tool — known as LocHap — tracks down local haplotype variant calling from deep sequence data to detect somatic mosaicism and cellular heterogeneity in individual samples, the researchers note. When applying this approach to whole-genome and exome sequences from normal samples and head and neck cancer samples, they found extensive local haplotype variation, though this variation and the accompanying somatic mosaicism was more pronounced in the tumor samples.