Scientists found a “don’t eat me” signal that cancer cells use to hide from the immune system, a new green nanotechnology could help clean up microplastics pollution from the world’s waterways, and a Chinese algorithm is coming for our jobs. Like, our jobs, specifically: It’s an algorithm that makes readable news stories out of complicated scientific studies. Until that happens, though, here’s this week’s (entirely human-made, handcrafted, organic and artisanal) roundup of the coolest scientific discoveries on the planet.
What is it? Scientists at the Stanford University School of Medicine discovered what they call a “don’t eat me” signal emitted by cancer cells — which, as its name suggests, discourages the immune system from eating, or otherwise attacking, the cancer.
Why does it matter? If the signal could be blocked or turned off, it would boost the immune system’s ability to fight cancer: Implanting human cancer into mice, the researchers demonstrated that blocking the signal — called CD24 — enabled the immune cells to go after the unwanted cells.
How does it work? Normally it’s the job of immune cells called macrophages to detect unwanted cells, “then engulf and devour them,” according to Stanford. In recent years, scientists have discovered proteins on cell surfaces that send the don’t-eat-me signal to the macrophages — fine enough for normal cells that shouldn’t be eaten, but it can also be used by cancer cells to essentially hide from the immune system. “You know that if cancers are growing in the presence of macrophages, they must be making some signal that keeps those cells from attacking the cancer,” said Amira Barkal, a Stanford MD-PhD student and the lead author of a new study published in Nature. “You want to find those signals so you can disrupt them and unleash the full potential of the immune system to fight the cancer.”
What is it? A team led by researchers at Australia’s University of Adelaide developed a system of “tiny coil-shaped carbon-based magnets” to break down the microplastics plaguing today’s waters.
Why does it matter? Earlier this year, a researcher from Bangor University in Wales said the problem of microplastics pollution was “absolutely everywhere” — in lakes, rivers, oceans and even groundwater around the world. The effects of microplastics on aquatic ecosystems and human health are still poorly understood, though early indications aren’t promising. Adelaide chemical engineering professor Shaobin Wang, senior author of the new study in Matter, said, “Microplastics absorb organic and metal contaminants as they travel through water and release these hazardous substances into aquatic organisms when eaten, causing them to accumulate all the way up the food chain.”
How does it work? Disposing of microplastics requires that they be broken down even further, via chemical reaction, into tinier pieces that can dissolve harmlessly in water — but many of the chemicals that can do this themselves require harmful pollutants to produce. The Adelaide researchers found a greener route by lacing carbon nanotubes with nitrogen, which helps generate the chemicals that decompose microplastics. They also injected the tubes with a tiny bit of manganese, making them slightly magnetic — and easy to collect and use again after they’ve finished a cleanup job.
What is it? Google’s artificial intelligence company, DeepMind, says it’s developed an algorithm that can alert doctors to signs of acute kidney injury in human patients up to 48 hours before it occurs.
Why does it matter? Acute kidney injury, or AKI, causes a buildup of waste products in the blood, can require timely dialysis to treat and is often linked to other health problems — according to the health news website STAT, the DeepMind researchers have been trying to help doctors keep “frail” patients from descending into a “life-threatening spiral.” Early signs of problems like AKI often get lost in a mountain of other patient data; having AI trained to search them out would enable doctors to swiftly treat the problem. “Giving doctors a head start on these major causes of patient deterioration that contribute to the deaths of literally hundreds of thousands of people every year could be absolutely transformative,” said DeepMind’s Dominic King.
How does it work? As described in a letter to Nature, study leaders trained an algorithm to look for patterns in the electronic health records of more than 700,000 patients in Veterans Affairs hospitals in the U.S. Though it ended up being able to predict 90% of AKI episodes up to 48 hours ahead, there remain some kinks to be worked out: particularly a high false-positive rate and the fact that the data the AI learned on skewed heavily toward male patients; only 6% of it came from women.
What is it? Meet LOCKR: the “first completely artificial protein switch that can work inside living cells to modify — or even commandeer — the cell’s complex internal circuitry,” developed by scientists at the University of Washington Medicine Institute for Protein Design and the University of California, San Francisco.
Why does it matter? “The ability to control cells with designer proteins ushers in a new era of biology,” said Hana El-Samad, a UCSF professor of biochemistry and the co-author of a pair of papers on the finding, both published in Nature. “In the same way that integrated circuits enabled the explosion of the computer chip industry, these versatile and dynamic biological switches could soon unlock precise control over the behavior of living cells and, ultimately, our health.”
How does it work? Proteins have a part in “almost all of the interesting things that happen in your body,” according to a summary of the findings in The Washington Post. Researchers have long tried to tweak them to better serve us, a tricky proposition because proteins don’t just have a single function — so reprogramming them could have unwanted effects. It’s better to just create a protein switch from scratch that can be programmed to do one thing and one thing only. LOCKR switches could direct cells to “modify gene expression, redirect cellular traffic, degrade specific proteins, and control protein binding interactions,” and could lead to new therapies for cancer and immune disorders. The researchers who created LOCKR hope it will be played around with by the scientific community at large: They’ve taken out patents for its commercial use, but made the DNA blueprints available to academics.
What is it? A team from a Chinese newspaper and Peking University has created Xiaoke, a “robot science reporter”: It’s an artificial intelligence algorithm that reads the abstracts of papers published in journals like Science and Nature, then summarizes them in easy-to-digest news stories.
Why does it matter? I mean, it’s not like I need a job or anything.
How does it work? In reality, the scientific-article-summarizing AI is not as alarming as it sounds: Its creators envision it not just as a way for newspapers to spread scientific news, but for researchers in China to get quick, accurate access to the latest advances described in English-language journals. After Xiaoke submits a first draft, the article are reviewed by scientists and newspaper editors who determine whether the content is good to go, or if it needs additional information. If you read Chinese, you can check out some of Xiaoke’s work here. See you next week. I hope!