‘One Of The More Powerful Antibiotics’
What is it? Researchers at the Massachusetts Institute of Technology used artificial intelligence to identify a molecule with powerful antibiotic properties — including against bacteria otherwise resistant to antibiotic treatment.
Why does it matter? The World Health Organization has described bacterial resistance to antibiotics as “one of the biggest threats to global health, food security, and development today.” While more bacteria are developing resistance to commonly used antibiotics, the development of new drugs tends to be slow and costly, as MIT medical engineering and science professor James Collins explained: “We’re facing a growing crisis around antibiotic resistance, and this situation is being generated by both an increasing number of pathogens becoming resistant to existing antibiotics, and an anemic pipeline in the biotech and pharmaceutical industries for new antibiotics.”
How does it work? As they write in Cell, Collins and colleagues developed machine-learning models capable of rapidly scanning a library of molecules and homing in on candidates with bacteria-killing potential. That’s how they identified the molecule they call halicin — named after Hal, the AI system in “2001: A Space Odyssey.” In the lab, halicin was found to be effective against antibiotic-resistant strains, including Clostridium difficile and Mycobacterium tuberculosis. Also, when tested on mice, it rid them of Acinetobacter baumannii — a bacteria that’s caused infection among U.S. soldiers stationed abroad. Collins said, “Our approach revealed this amazing molecule which is arguably one of the more powerful antibiotics that has been discovered.”
Breaking News: Bendable Energy Storage
What is it? Scientists at University College London and the Chinese Academy of Sciences developed a bendable supercapacitor that “safely stores a record-high level of energy for use over a long period.”
Why does it matter? As UCL noted in a release, “high-powered, fast-charging supercapacitors” have previously faced a roadblock: They typically lack the ability to store large amounts of energy in a small space. This new development could be used in phones, wearable technology, electric vehicles and other applications. Zhuangnan Li, first author of a new paper in Nature Energy, said, “Our new supercapacitor is extremely promising for next-generation energy storage technology as either a replacement for current battery technology, or for use alongside it, to provide the user with more power.”
How does it work? The device uses an electrode material made from multiple layers of the wonder material graphene, enabling it to store a charge more efficiently. Li said, “We designed materials which would give our supercapacitor a high power density — that is how fast it can charge or discharge — and a high energy density, which will determine how long it can run for. Normally, you can only have one of these characteristics, but our supercapacitor provides both, which is a critical breakthrough. Moreover, the supercapacitor can bend to 180 degrees without affecting performance and doesn’t use a liquid electrolyte, which minimizes any risk of explosion and makes it perfect for integrating into bendy phones or wearable electronics.”
Accepting The Charges
What is it? Antibiotics development isn’t the only pipeline that AI is helping to unclog: See also battery performance. At Stanford, researchers developed a machine-learning model that could “supercharge” battery development for electric vehicles.
Why does it matter? In this case, the clog in the pipeline is testing: As Stanford explains in a news release, “At every stage of the battery development process, new technologies must be tested for months or even years to determine how long they will last.” The AI methods developed by researchers Stefano Ermon and William Chueh could cut testing times by 98%, Ermon said: “At every stage of the battery development process, new technologies must be tested for months or even years to determine how long they will last. With AI, we’re able to quickly identify the most promising approaches and cut out a lot of unnecessary experiments.” The goal is EV batteries with charging times comparable to what it takes to fill up your tank with gas.
How does it work? Though they say their AI method could be used at various points in the battery-research process, the Stanford scientists focused on one facet for a new study published in Nature: “finding the best method for charging an EV battery in 10 minutes that maximizes the battery’s overall lifetime.” They developed a machine learning model to predict “how batteries would respond to different charging approaches,” and found that by reducing both the number of trials and how long each took, they could reduce the testing process drastically: from about two years to 16 days.
What is it? Researchers at the University of Southern California developed a tool called SEER that enables scientists to “peer more deeply and clearly into living things.”
Why does it matter? A persistent challenge in medical imaging is that bodies, human and otherwise, tend to be — well — pretty opaque: Doctors have to sort through a lot of visual information. SEER, which combines imaging with mathematical computations, “provides greater clarity and works up to 67 times faster and at 2.7 times greater definition than present techniques,” according to USC. The researchers who developed it say SEER, which could be used in a smartphone app, has potential applications beyond medical science — say, in food safety or identifying counterfeit currency.
How does it work? SEER, which stands for “spectrally encoded enhanced representations,” starts with fluorescence hyperspectral imaging, or fHSI. Researchers use fHSI to “differentiate colors across a spectrum, tag molecules so they can be followed, and produce vividly colored images of an organism’s insides.” But it yields so much data that gathering and processing images slows things down dramatically, which impedes doctors’ ability to make decisions faster. SEER uses computer calculations to speed up the process. USC’s Francesco Cutrale, lead author of a new study in Nature Communications, said, “There is a number of scenarios where this after-the-fact analysis, while powerful, would be too late in experimental or medical decision-making. There is a gap between acquisition and analysis of the hyperspectral data, where scientists and doctors are unaware of the information contained in the experiment. SEER is designed to fill this gap.”
Making It Of-fish-al
What is it? Working with colleagues in China, scientists at UCLA developed an antifreeze coating after drawing inspiration from animals that have a lot of experience when it comes to not freezing: fish that swim through icy waters near Antarctica.
Why does it matter? The coating — described in Matter and characterized by UCLA as “the first material that prevents ice formation by acting on three distinct aspects of ice formation” — could be used to prevent ice from forming on airplane wings, pipes and other outdoor equipment. UCLA materials science and engineering professor Ximin He said, “Ice formation starts from nucleation, when a small seed crystal of ice first forms, before it grows and then finally adheres to a surface. While there are anti-ice solutions out there, they’re designed to tackle only one of these three aspects of this complex process, or they only work on certain types of surfaces. This new coating is an all-in-one solution to prevent ice formation on many different surfaces, from plastics to metals to ceramics, and under different conditions.”
How does it work? The key ingredient of the gel that He and colleagues came up with is the silicone-based polymer polydimethylsiloxane, used in contact lenses, lubricants and “other applications that require some slipperiness.” It forms a thin coat on surfaces where it’s sprayed that, as mentioned above, works in three ways: by lowering the freezing temperature of water on the surface, delaying the formation of ice crystals and preventing ice from sticking to the surface. The substance mimics the molecular structure of proteins that scientists identified decades ago in the blood of some species of Antarctic fish."