| Brought to you by The 5th Innovation Forum on Intelligent Computing |
 |
|
|
|
|
| 26 September 2025 |
| Today’s Deep Dive asks: What’s in a (species) name? But first, catch up on the latest science news, including the pros and cons of standing out and why we sigh. |
|
|
|
| Animals | News from Science |
| How fungus-farming termites control weeds |
| More than 50 million years before humans plowed their very first field, termites began farming fungi inside their nests for food. And just like human farmers, termites must contend with “weeds,” in the form of unwanted fungi that can spoil their crop. In this week’s Science, researchers report one way that termites keep their crop in good order: by burying noxious fungi within soil that contains antifungal microbes.
The termites in the new study, a southwest Asian species called Odontotermes obesus, prepare their “fields” by bringing bits of leaves into the nest. The worker termites then chew the leaves into tiny bits and stuff them into special cavities that are at the right temperature and humidity for a fungus called Termitomyces to thrive. As the white fungus grows on the leaf matter, called comb, the termites continually reap and eat it.
Other kinds of fungi can compete with Termitomyces. The researchers were curious how the termites keep these unwanted fungi in check. So, they dug up comb and termites and brought them into the lab. The team gave termites both healthy comb and comb on which they placed a common weedy fungus, Pseudoxylaria. The termites buried the contaminated comb but not the healthy comb. Further experiments showed that microbes in the soil combat the unwanted fungus.
The team is now studying how the microbes specifically inhibit fungi. They also hope to generate a little more public respect for termites. “As fungus-growing termites remain underground, and notoriously difficult to work with, very little is known about their unique biology,” noted evolutionary biologist Rhitoban Raychoudhury. “We hope that people realize that these out-of-sight insects also have very interesting lifestyles.” |
| Read the SCIENCE PAPER and RELATED PERSPECTIVE |
|
|
|
|
|
| Ecology | Science |
| Is it better to blend in or stand out? |
 |
| Warning colors didn’t protect this butterfly. Stanislav Harvancik |
| The 2004 classic Mean Girls famously shows that the animal kingdom and high school aren’t so different. The creatures in each need to make a tough choice for survival amidst top predators: blend in or stand out? There are merits to both approaches, suggests new research.
To study how insects avoid predation, researchers placed more than 15,000 paper moths in forests across six continents, each pinned with mealworm bait. Some of the fake moths were camouflaged in bark-colored brown, and some had warning patterns of bright orange or turquoise. Then the team monitored how often birds ate each kind of moth.
It turned out that successfully avoiding predation depended on the surrounding ecology. Camouflage was a good strategy in low-light conditions or where predators were common. Warning colors were more successful when there were fewer predator species around, meaning birds didn’t test out a brightly colored snack out of necessity. In general, both strategies thrived when the surrounding animals tried the opposite tactic; in other words, camouflage worked best when most other creatures had warning colors, and being bright was successful when nearby prey blended in.
While the authors wrote that “there was no overall ‘best’ strategy,” camouflaging was likely more vulnerable to ecological change and therefore more often lost and regained throughout evolutionary history. |
|
|
|
|
|
| Physiology | Science Advances |
| *Sigh*—but why? So you can breathe easy |
| Sighing is an essential human reflex—which is why we do it roughly once every 5 minutes. And there’s more to it than simply forcing open any of the little air sacs in your lungs that have collapsed. According to new research, sighing helps rearrange molecules in the mucusy layer inside the alveoli that makes contact with the air, which in turn helps prevent the alveoli from collapsing when you exhale.
Thanks to mechanical ventilators, we know that if a person only breathes the ordinary amount—exchanging about 10% of the air in their lungs—the lungs become harder to inflate over time. Maria Clara Novaes-Silva and her colleagues wanted to know exactly why that is, and what it is about sighing that ‘resets’ this. So, they took a super close look at what happens to the part of the lungs that’s actually in contact with air.
“Inside of our alveoli, we have this very thin liquid layer, and this creates a liquid–air interface,” Novaes-Silva explained to Science Podcast Host Sarah Crespi. “We are constantly expanding and compressing this area.” This liquid is a mixture of lipids and proteins that forms a multilayered film.
When you breathe normally, you stretch this film a little. But when you sigh, you breathe in more than twice as much air—and that quickly stretches the film, which then compresses as you exhale. Using artificial pulmonary fluid, Novaes-Silva and her colleagues showed that this process redistributes lipids: moving tight-packing saturated fats to the top, air-contacting layer and looser-packing unsaturated lipids to the lower layer. Overall, this makes the film easier to stretch—making inhalation easier—as well as more resistant to compression, making alveoli more resilient to collapse. |
|
|
|
|
|
 |
|
|
| Lab to market: Turning research into real-world impact |
| Join Science Webinars October 3 for a live talk with Jay Keasling, renowned synthetic biologist and entrepreneur. Learn how Keasling is developing life-saving drugs, sustainable chemicals, and building companies that bring these solutions to market. |
|
|
|
|
|
|
|
| Deep Dive |
 |
| The Yunxian 2 cranium that some may claim should be labeled Homo longi Gary Todd via Wikimedia Commons | CC0 |
| What’s in a name? Ask a Denisovan |
| Michael Price, Deputy News Editor, Science |
| Want to start a fistfight at an anthropology conference (or at least a polite verbal tussle)? Ask which species Neanderthals and their close cousins, the Denisovans, belong to.
The definition of a species in the study of evolution has always been a slippery and imprecise one. You may have learned in school that two animals belong to the same species if they can produce viable offspring, but that’s an oversimplification. After all, coyotes and wolves can have babies that have babies, but few would argue that the two canids are the same species. In truth, the boundaries between species are often messy, contentious and, ultimately, arbitrary.
Regarding Neanderthals in the above question, there are a few different camps the answers might come from. Some might go with Homo neanderthalensis, first proposed in 1863, named after Germany’s Neander Valley where the first identified Neanderthal was found. Others argue Neanderthals are a subspecies of our own species, Homo sapiens, making them Homo sapiens neanderthalensis.
No less contentious are the proposed designation for Denisovans. No formal species name has been given to this hominin, which was first identified in 2010 based on the analysis of mitochondrial DNA sequenced from a finger bone found in Siberia’s Denisova Cave. But one proposed candidate is Homo longi, a species name proposed for a skull found in Harbin, China, that was described in 2021. Researchers argued its morphological characteristics were distinct enough from other known hominins to warrant a separate species name. Then, earlier this year, ancient proteins confirmed the Harbin skull was a Denisovan. Now, a paper out this week in Science argues, based on morphological analysis, that a 1-million-year-old Chinese skull known as Yunxian 2, previously classified as Homo erectus, belongs to H. longi, too . And based on physical similarities between known Denisovans, the Harbin skull, and Yunxian 2, the authors argue that Denisovans most likely belong to the H. longi clade.
So, does that mean Denisovans should now be called Homo longi? According to the rules set down by the International Code of Zoological Nomenclature, there’s a good case to be made that they should. After all, according to the ICZN’s so-called Principle of Priority, “the valid name of a taxon is the oldest available name applied to it, unless that name has been invalidated or another name is given precedence.” So, assuming you buy the argument that modern humans, Neanderthals and Denisovans deserve to be classified as separate species in the first place, then there’s a good argument that Denisovans should now be considered H. longi.
Still, there’s room for dissent. Some have argued that the Harbin skull may not accurately represent the breadth of Denisovan diversity, and just because the Harbin Denisovan can be called H. longi doesn’t mean that all Denisovans should be lumped under the same taxonomic category. In this view, there may yet be other hominins currently categorized under the broad term Denisovan that deserve entirely different species names.
Time and debate and flurries of papers will eventually settle this issue. Punches may or may not get thrown. |
|
|
|
|
|
| Letai for NCI? |
| President Donald Trump is expected to tap Anthony Letai, a highly regarded Harvard Medical School oncologist and basic scientist, to lead the National Cancer Institute. He “brings a strong basic science background, which is very important for NCI,” said former NCI director Monica Bertagnolli. |
| Read more at ScienceInsider |
|
|
|
| Turbulence reimagined |
| Physicists’ new model to explain turbulence could help engineers and pilots figure out how to make flights smoother. “Airplane design is going to benefit,” one expert said. “The better the model, the more it captures of the particular turbulent field, then the better the forecast, which is what the pilot is going to use,” added another. |
| Physical Review Research Paper | Read more at The New York Times |
|
|
|
| The importance of interpretability |
| To truly understand LLMs, experts need to examine their training, not just how they behave at the end of it, according to Naomi Saphra. “We don’t know what makes a language model tick,” she said. “If we have these models everywhere, we should understand what they’re doing.” |
| Read more at Quanta Magazine |
|
|
|
|
|
| Subscribe to News from Science |
| Subscribe for unlimited access to authoritative news on science research and policy |
|
|
|
|
|
|
|
|
| Brought to you by The 5th Innovation Forum on Intelligent Computing |
 |
|
|
|
|
|
|
|
|