Sciber

"The scientific activities of the Tara Oceans expedition, led by EMBL senior scientist Eric Karsenti, present an unprecedented effort that resulted in 35,000 samples containing millions of small organism collected in more than 210 ocean stations, chosen for their climatic significance or biodiversity. Putting to work its extremely advanced microscopy facility, analysing the genetic sequences of all organism with sophisticated bioinformatics tools, and explaining the story of life through surprising opportunities in evolutionary biology, EMBL is putting the crowning analysis on top of one of the most ambitious projects of our time". More @ EMBL

"In general, Majumdar and Schehr believe, systems in the Tracy-Widom universality class exhibit one phase in which all components act in concert and another phase in which the components act alone." News article @ WIRED

"Professor Marcus Feldman's lab has devised a computer model that could help solve a long-standing mystery over why the introduction of new tools in prehistoric societies sometimes comes in periodic bursts."Biologists trace how human innovation impacts tool evolution

"Using chemogenetic tools to spur the brain into action. [...] Less than a decade ago, such precise control over neuronal activity in a dish, let alone in a living brain, was impossible. "Remote Mind Control | The Scientist Magazine®

In 1855, Charles L. Dodgson became the mathematical lecturer at Christ Church College in the University of Oxford, UK. His job was to prepare Christ Church men (for it was all men) to pass examinations in mathematics. Dodgson (1832–98) would go on to publish Alice's Adventures in Wonderland (1865) and Through the Looking-Glass (1871) under the pen name Lewis Carroll, but he also produced many pamphlets and ten books on mathematical topics. In some of these, he exhibited unusual methods — for rapid arithmetic, for example. Others featured innovative ideas that foreshadowed developments in the twentieth century, for instance in voting theory". Full review @ Nature

recent puzzle in the big data scientific literature is that an increase in explanatory variables found to be significantly correlated with an outcome variable does not necessarily lead to improvements in prediction. This problem occurs in both simple and complex data. We offer explanations and statistical insights into why higher significance does not automatically imply stronger predictivity and why variables with strong predictivity sometimes fail to be significant. We suggest shifting the research agenda toward searching for a criterion to locate highly predictive variables rather than highly significant variables. We offer an alternative approach, the partition retention method, which was effective in reducing prediction error from 30% to 8% on a long-studied breast cancer data set. Full paper @ PNAS