Tree of life gains new genomic branches

In a noteworthy feat, the number of decoded genomes grew by nearly 8,000 species in a single day. The newly available genomes, published in Nature Microbiology in September 2017, mainly represent new microbial species. Their addition bumps up the number of decoded genomes by 10%. The Australian research team, based at the University of Queensland, is interested in microbial diversity and evolution. In particular, they wanted to establish better genomic representation among species that are not easily isolated in the lab and grow poorly or not at all in culture. They used advances in computational techniques to create metagenome-assembled genomes from high-diversity environments. The new genomes provide the first representatives of 17 bacterial and 3 evolutionarily distinct archaeal groups. In addition, the new genomes increase representation for groups such as Actinobacteria, a group of soil- and waterborne microbes that have yielded drugs such as antibiotics, antivirals, and chemotherapy agents. Investigator Donovan Parks remarked in a prepared statement that this research highlights the substantial inroads that have been made into the vast diversity of phylogenetic and metabolic life. However, the findings represent only a small fraction of the environmentally derived genome fragments available for study. That leaves a large swath of microbial diversity remaining to be sequenced. Parks predicted that new environmental samples deposited in public repositories would add many tens of thousands of additional microbial genomes in the near term. “Constructing a comprehensive genomic repository of microbial diversity lays the foundation for furthering our understanding of the role of microorganisms in critical biogeochemical and industrial processes,” he said. —Karyn Hede, News Editor

Public tussle erupts over Craig Venter's latest claim to decode faces from DNA

Taking facial recognition to a level approaching science fiction, genome pioneer Craig Venter and his colleagues have raised public alarm bells over a claim to be able to identify human faces based solely on DNA sequences. Rarely do technical scientific reports generate the kind of public scrutiny that the announcement by Venter and his colleagues at Human Longevity, San Diego, CA, has drawn. The company's recent effort to use computational modeling of single-nucleotide polymorphism-based data to predict features such as skin and eye color, height, age, and facial structure, was published online in September 2017, in the Proceedings of the National Academy of Sciences. The researchers claimed that their algorithm could correctly identify an individual out of 10 randomly selected from their database 74% of the time. Within hours of its public availability, a researcher who had reviewed an earlier version of the paper, Yaniv Erlich, published a scathing critique of the study methodology on bioRxiv, the open-access preprint server for biology. Venter and colleagues rebutted the critique, also on bioRxiv. Meanwhile, the scientific spat spilled over into social media, where followers on Twitter weighed in. While the debate may seem to be an academic argument, or even a personal vendetta, some scientists point out that this effort, along with Venter's high public profile, could skew the debate over genetic privacy and the perceived risk that a person's genome could be used to match a DNA sample to their face. Only time will tell which side ends up with egg on their face. —Karyn Hede, News Editor