In this week's Nature Communications, a team from the University of Oxford presents a new computer program that can detect antibiotic resistance in bacteria by analyzing the microbes' genomes. Called Mykrobe Predictor, the program is capable of detecting both species and resistance profiles of bacterial samples isolated from patients, and was used to predict resistance to 12 antibiotics in Staphylococcus aureus and Mycobacterium tuberculosis samples. GenomeWeb has more on this study here.
Meanwhile, in Nature Genetics, a group of US and Swedish scientists publish a study revealing the genetic mechanism responsible for a horse coloration pattern known as dun, which is characterized by a handful of dark markings on a relatively light-colored coat and believed to be the ancestral coloration in horses. Using SNP genotyping, comparative genomics, and microscopy, the investigators found that dun patterning is caused by the asymmetric expression of a gene called TBX3 that codes for a pigment-squelching transcription factor. They further discovered that a non-dun mutation was already present in ancient horses and likely selected for by humans during domestication.
Also in Nature Genetics, Stanford University researchers report on the use of density-based clustering techniques using cancer-, mutation type-, and gene-specific mutation models to identify regions of recurrent mutation in 21 cancer types. They found recurrent mutations across a variety of coding and non-coding elements in up to about 15 percent of specific tumor types, with significantly mutated regions in both well-established cancer drivers and in new genes and functional elements. Overall, their study revealed "a rich spectrum of coding and non-coding elements recurrently targeted by somatic alterations that complement gene- and pathway-centric analyses."