Sounding the alarm on the private sector’s increasing dominance of artificial intelligence (AI), researchers at MIT this week warn that with industry poised to gain control over the technology’s future, academia—and public interest—could get left in the dust. In a perspective aimed at policy makers, they detail the brain drain from universities to private companies over the last two decades as well as the explosion of funding in the private sector and the industry’s growing share of publications and AI models. The goal, however, should not be to steer more research back toward academia they say, but simply to find ways of enshrining expertise in public settings so that public interest is not lost or forgotten in the gold rush. “The goal should be to ensure the presence of sufficient capabilities to help audit or monitor industry models or to produce alternative models designed with the public interest in mind,” they write. Science
The need for personalized medicine in many human diseases is driven by the simple fact that diseases themselves vary from person to person. The complex pathologies of inflammatory autoimmune diseases like rheumatoid arthritis, Crohn’s disease, and ulcerative colitis, for instance, are marked by the non-uniform involvement of thousands of genes all over the body—meaning the exact genes and their roles vary from person to person. That makes the diseases hard to understand and harder still to treat. But researchers at the Karolinska Institute in Stockholm, Sweden, have discovered how different genome-wide changes in these diseases may get switched on or off by the same pro- or anti-inflammatory upstream regulators. Targeting these upstream regulators may give doctors an easier way to design personalized treatments for patients to control these diseases. Cell Reports Medicine
Researchers at the University of Toronto have found a potential new treatment for Parkinson’s disease in a very unusual way. Using the IBM Watson for Drug Discovery AI program, they searched existing drugs for the ability to induce “mitophagy” and clear damaged mitochondria organelles from the brain cells, a normal process that’s known to go awry in Parkinson’s, leading to the loss of neurons and the classic motor symptoms of the disease. Using a semantic similarity search, identifying the compound not based on data alone but on the words other researchers used to describe it, they identified an existing cholesterol-lowering drug on the market called probucol and showed in two different animal models of Parkinson’s disease that it could improve motor function and slow neuron loss. Which just goes to show, words still matter. PLOS Biology
Per- and polyfluoroalkyl substances (PFAS) are common manmade chemicals added to soaps, cosmetics, cookware, food packaging, and many other everyday products. However, they are endocrine-disruptors that can interfere with hormones in the body, and PFAS exposure has been linked to high cholesterol, thyroid disease, cancer, and pregnancy-induced hypertension, among other health problems. Now researchers at the University of Florida may have found an unexpected potential source of PFAS chemicals in groundwater: toilet paper. Obtaining packaged toilet paper from North America, South and Central America, Western Europe, and Africa, they looked for 34 known PFAS chemicals and found six, the most abundant being a chemical commonly used in paper mills. They also found the same PFAS chemicals in sewage sludge from eight wastewater treatment plants in Florida. “Our results suggest that toilet paper should be considered as a potentially major source of PFAS entering wastewater treatment systems,” they write. All around the world. Environmental Science & Technology Letters
Subscribe to our free weekly newsletter.