A magnetar is born, and Alzheimer's reveals a new cell killer

Not a steady blaze, but a rhythmic brightening and fading that sped up as the weeks passed, like a top slowly winding down [4]. Graduate student Joseph Farah at Las Cumbres Observatory brought the data to UC Berkeley astrophysicist Dan Kasen, who had published a theory about this exact signal sixteen years earlier and never had confirmation. His 2010 paper, cowritten with Lars Bildsten, proposed that the universe's most brilliant supernovae stay bright long after they should fade because the dying core collapses into a magnetar: a neutron star about ten miles across, spinning more than a thousand times per second, with a magnetic field a hundred to a thousand times stronger than a typical pulsar. Spinning at that rate, it pumps energy into the surrounding explosion like a battery, keeping it blazing. SN 2024afav's chirp matched the theory exactly, and the precise timing required Einstein's general relativity to decode, making this the first time general relativity has been used to explain how a supernova actually works. The results were published in Nature [4].

It works like this: toxic proteins accumulate inside a neuron, the cell's nucleus begins to shrink, and eventually the genetic material inside collapses and disintegrates, killing the cell [3]. Scientists had long identified other ways neurons die, including apoptosis, but those mechanisms never fully explained the scale of neuron loss seen in Alzheimer's brains. Working with the UK Dementia Research Institute, the King's team analyzed 3,000 brain cells collected from 28 people who had died with either frontotemporal dementia or advanced Alzheimer's disease. They found signs of karyoptosis in 35 percent of cells from the frontal cortex of people with Alzheimer's, compared to 15 percent in healthy older adults. They also identified a group of molecular switches, proteins called kinases, that appear to control the process. Force those proteins to clump, as Alzheimer's pathology does, and karyoptosis begins. "This study is the culmination of a 10-year journey at King's," the researchers wrote, "from when we first identified karyoptosis in a relatively rare disease to discovering that it is a common feature of dementias which affect millions of people." Published in Nature Communications [3].

Nothing had worked. Their average age was 69, and for most of them, a knee replacement was what remained [25]. The procedure the Berlin team tried instead is called genicular artery embolization, or GAE. It targets the abnormal blood vessels that grow around an arthritic joint and fuel the pain-sensing nerves surrounding it. Close off those vessels, and the nerves quiet down. The key change here was the material used to do it: previous GAE trials used antibiotics, which raised concerns about resistance and inflammation. This team used resorbable microspheres, microscopic gelatin beads that dissolve within hours, breaking the cycle of inflammation and pathological nerve growth without permanently altering the knee. "By embolizing the pathological vessels, we're able to normalize the vessel structure, and in turn, the neuronal structure of the knee," said radiologist Florian Nima Fleckenstein. At 12 months, average pain was 3 out of 10. Scores for daily activity, sports, recreation, and quality of life all improved substantially [25].

After several weeks, the students texting real strangers reported feeling significantly less lonely. The students chatting with the AI did not. This wasn't because the AI was bad at its job. The chatbot was warm, attentive, and responsive, genuinely good at its role. The gap seemed to come from something else. The students texting strangers knew they had formed an actual connection with another person who had chosen to respond to them. That reciprocity, a human somewhere deciding to write back, was something the AI could not replicate, and the students seemed to sense its absence. "Building real social connections requires more than simply receiving simulated empathy," the researchers concluded. The study was published in the Journal of Experimental Social Psychology [38].

The power numbers are real: the two test sections together generate 1.6 megawatts and cover an area about one and a half football fields [126]. But the water numbers surprised everyone. Shade from the panels reduced water evaporation by up to 70 percent. Aquatic weed and algae growth dropped by up to 85 percent. That second finding matters more than it sounds, because canal operators spend considerable money managing those weeds through herbicides and mechanical clearing every season. Lead researcher Brandi McKuin at UC Merced modeled what would happen if all 4,000 miles of California's major open canals were covered: 13 gigawatts of generating capacity, roughly half the new solar California needs to hit its 2030 renewable targets, and 63 billion gallons of water saved per year, enough to supply 2 million people or irrigate 50,000 acres of farmland [126].

Their review was published in The Lancet [69]. The conclusions were unambiguous. mRNA vaccines are safe and highly effective at preventing the spread of infectious diseases like COVID-19. They do not alter human DNA. They deliver temporary instructions that allow cells to mount an immune response, then disappear. The review confirmed they are safe for children and pregnant women, that boosters extend protection, and that the same platform shows real promise for cancer treatment, training the immune system to recognize tumor cells. Lead author Anna Blakney at the University of British Columbia described it as "an extraordinary amount of scientific evidence" accumulated across billions of real people. The paper also named the side effects honestly. Some people experience allergic responses. A small subset of young men have developed myocarditis. The authors were direct: both are rare, and neither comes close to outweighing the benefits [69].

Small, dense particles are the ones that get trapped in artery walls and build into plaque. Two people can have identical LDL numbers and very different particle counts [1]. ApoB, or apolipoprotein B, counts the particles directly. There is one apoB molecule on every harmful cholesterol particle in the blood, so the apoB test is a precise census of what's actually circulating. A study from Northwestern University's Feinberg School of Medicine, published in JAMA, found that using apoB to guide treatment decisions would prevent more heart attacks and strokes than using LDL, and that switching to apoB-guided care would be cost-effective for the U.S. healthcare system. "We found that apoB testing to intensify cholesterol-lowering medication would prevent more heart attacks and strokes than current practice, and that these health benefits were achieved at a cost that represents good value for U.S. healthcare payers," said lead author Ciaran Kohli-Lynch [1]. It's the first study to show apoB is also economically viable at scale.