Illuminating math and science. Supported by the Simons Foundation. 2022 Pulitzer Prize in Explanatory Reporting. www.quantamagazine.org
In 2023, tiling hobbyist David Smith found a single tile that resisted repetition at least as far as his attempts could take him. Could this be the elusive, aperiodic monotile? Tune in to this week’s “The Joy of Why.” listen.quantamagazine.org/jow-312-s
0 replies
1 reposts
4 likes
Not every theoretical discovery has practical applications. For some, a victory over mathematical impossibility is its own reward.
www.quantamagazine.org/amateur-math...
1 replies
0 reposts
13 likes
How exactly do magnetic fields shape galaxies? It’s not a question you can answer head-on. Instead, astrophysicist Susan Clark takes the circuitous route: She finds clues in the way light passes through the cosmic force fields. www.quantamagazine.org/tracing-the-...
1 replies
5 reposts
9 likes
What’s the worst centrally symmetric convex shape for packing the plane? A hexagonal grid of circles covers 90.69% of the plane when optimally packed. An octagon with rounded corners covers a smidgen less: 90.24%. But is this shape the worst one?
www.quantamagazine.org/why-is-this-...
0 replies
0 reposts
4 likes
Read the full extended feature here: quantamagazine.org/how-ai-revol...
15/15
0 replies
2 reposts
2 likes
Still, hesitancy pervades. If science uses deep learning tools that show solutions and no process, is it really science? If we give up on understanding nature’s processes, then artificial intelligence will not just have changed science — it will have changed scientists too. 14/15
1 replies
0 reposts
3 likes
What’s next? It’s time to consider the intricate environment of the cell.
Launched in spring 2024, AlphaFold3 and RoseTTAFold All-Atom use deep learning algorithms to predict the structures of proteins bound to each other, DNA, RNA and other small molecules.
13/15
1 replies
0 reposts
0 likes
“Some people are overconfident — like, way overconfident — in what these deep learning models can do,” said Lauren Porter, an investigator at NIH. “We should use these deep learning models for as much as we can, but we also need to approach them with caution and humility.” 12/15
1 replies
0 reposts
2 likes
Biologists are confined to studying the proteins already invented by nature. But David Baker envisions a future in which they could design novel proteins — ones crafted specifically to harness sunlight, break down plastic, or form the basis of drugs or vaccines.
11/15
1 replies
0 reposts
0 likes
Biologists begin to combine AlphaFold2’s predictions with advanced electron microscope techniques like cryo-EM, which flash-freezes biological samples and bombards them with electrons.
AlphaFold2 has “turbo-boosted” cryo-EM, says systems biologist Mohammed AlQuraishi.
10/15
1 replies
0 reposts
0 likes
In 2020, Google DeepMind submits their improved predictions to CASP, and the world of protein science changes forever: AlphaFold2 accurately predicts the structure of a protein from its amino acid sequence.
Protein scientists have one question in mind: What now?
9/15
1 replies
0 reposts
0 likes
Google DeepMind reboots their algorithm and designs the transformer, a type of deep learning.
Things begin to click into place. The further they hone it, the more they notice enormous leaps in the efficiency and accuracy of AlphaFold’s protein folding predictions.
8/15
1 replies
0 reposts
0 likes
Google DeepMind wins, but not by a landslide. They return to the drawing board.
With original insights from a diverse background in physics, chemistry, biology and computation, John Jumper soon leads the team.
7/15
1 replies
0 reposts
1 likes
When John Jumper joins Google DeepMind in 2017, they’re already developing AlphaFold, a deep learning algorithm for predicting protein structure.
In 2018, the team enters AlphaFold into CASP. Stakes feel high. They consider competing anonymously to avoid humiliation.
6/15
1 replies
0 reposts
0 likes
It’s 2016. Google DeepMind, an AI team based in London, uses a deep learning algorithm to beat a human champion at an ancient board game called Go.
Scientists begin to wonder: Could deep learning be the missing puzzle piece of the protein problem?
5/15
1 replies
0 reposts
0 likes
A forum emerges: John Moult co-founds the Critical Assessment of Structure Prediction, or CASP, a competition for best approaches to the protein folding problem. Meetings are unconventional: Attendees stamp their feet if they dislike an entry.
4/15
1 replies
0 reposts
2 likes
As protein science expands, problems arise. Experimentalists work precisely but slowly; computationalists work quickly but are too removed from biophysical realities.
John Moult sets out to find a way to bring the best of both approaches together.
3/15
1 replies
0 reposts
0 likes
It’s the early 1960s. Biologists are growing proteins into crystals, bombarding them with X-rays and measuring how the rays bend — a technique known as X-ray crystallography.
From there, they create ball and stick models. Every finished model represents years of work.
2/15
2 replies
0 reposts
1 likes
Proteins do it all. Hemoglobin ferries oxygen around the body. Keratin structures hair, nails and skin. Insulin helps glucose convert into energy.
The fold of a protein is critical to its function. Yet no one really knows specifically how protein folding happens.
1/15
1 replies
6 reposts
16 likes
Scientists have the tools to construct a human cell atlas: a holistic reference map of a human being that could help to pin down which specific cells in which tissues go wrong to cause diseases. Listen to this week’s episode of the Quanta Science Podcast:
listen.quantamagazine.org/qsp-243-s
0 replies
0 reposts
3 likes
According to one researcher, Google DeepMind’s creation of AlphaFold is “the biggest ‘machine learning in science’ story that there has been.” Read the full story of AlphaFold’s creation and what it means for the future of AI in biology.
www.quantamagazine.org/how-ai-revol...
0 replies
2 reposts
5 likes
Sometimes, a problem is hard to solve no matter how you approach it. Researchers recently proved that multicalibration, an approach for ensuring fairness in algorithms, can also help point out instances of computational hardness. www.quantamagazine.org/the-question...
0 replies
1 reposts
4 likes
Many of the most accurate truths arising from science are in the domain of physics. Why? Physics, at its core, is all about the art of asking very, very simple questions. Learn more on “The Joy of Why.” listen.quantamagazine.org/jow-311-s
0 replies
2 reposts
5 likes
Innate biological clocks use light levels to determine how organisms grow, sleep, move and reproduce. New research suggests that similar chronobiology may be responsible for the spectacular synchronization of fruiting among millions of European beech trees. www.quantamagazine.org/across-a-con...
0 replies
2 reposts
5 likes
The grade school lesson that water freezes at 0 degrees Celsius is seldom true. To get started, fire needs a spark. And ice needs a nucleus. With new tools and inspiration from nature, scientists are finding the best ways to make ice. www.quantamagazine.org/the-enduring...
0 replies
4 reposts
10 likes
The brainstem controls your heart rate, breathing and, according to a new paper, inflammation throughout your body. This research suggests potential new treatments for inflammatory disorders like IBS, rheumatoid arthritis, and diabetes. www.quantamagazine.org/the-brainste...
0 replies
1 reposts
5 likes
Everything is made of stardust. Gamma-ray bursts may help physicists learn how each element in the periodic table is born in the cosmos. Listen to the new episode of the Quanta Science Podcast. listen.quantamagazine.org/qsp-242-s
0 replies
1 reposts
5 likes
Friedrich Nietzsche is reputed to have said, “Those who were seen dancing were thought to be insane by those who could not hear the music.” In the same sense, our world is governed by the silent music of computation. www.quantamagazine.org/computation-...
0 replies
1 reposts
5 likes
What does Jupiter’s Great Red Spot have in common with consciousness? A new theoretical framework suggests that scientists think about these emergent phenomena as a kind of “software in the natural world.” www.quantamagazine.org/the-new-math...
0 replies
2 reposts
7 likes
Dive into extra dimensions on our website: www.quantamagazine.org/tag/extra-di...
0 replies
0 reposts
3 likes
Many artists include their own interpretations of the fourth dimension in their work, including Pablo Picasso’s 1910 Portrait of Daniel-Henry Kahnweiler.
|
1 replies
0 reposts
2 likes
Salvador Dalí showcases a tesseract in his 1954 painting, Crucifixion (Corpus Hypercubus).
|
1 replies
0 reposts
3 likes
We can get some intuition for what a tesseract looks like by unfolding its three-dimensional boundary to obtain its eight boundary cubes.
|
1 replies
0 reposts
3 likes
To visualize a 4D shape, we can build up to it. Just as a square in two dimensions is bounded by four one-dimensional line segments, and a three-dimensional cube is bounded by six squares, the boundary of a tesseract — a four-dimensional cube — is defined by 8 three-dimensional cubes.
|
2 replies
0 reposts
8 likes
Mathematicians and artists have engaged in the mental exercise of imagining extra dimensions. What would a fourth dimension — somehow perpendicular to the familiar three — look like?
|
2 replies
1 reposts
10 likes
Expander graphs can be used to attack a wide variety of problems, from modeling neurons in the brain to generating error correcting codes. A recent proof shows hints at how structure works in some of these networks.
www.quantamagazine.org/in-highly-co...
0 replies
2 reposts
4 likes
On Tuesday, an FDA panel recommended against using MDMA to treat PTSD. The psychiatric future of psychedelics is a rapidly evolving landscape, with plenty of potential pitfalls and opportunities worth exploring. Listen to “The Joy of Why” to learn more. listen.quantamagazine.org/jow-310-s
0 replies
1 reposts
3 likes
Balon, a recently discovered “hibernation factor” protein, is shockingly easy to come by in nature. Present in 20% of all cataloged bacterial genomes, the protein establishes dormancy by throwing sand in the gears of cellular machinery that build proteins. www.quantamagazine.org/most-life-on...
0 replies
1 reposts
5 likes
The innocuous-sounding problem of distinguishing two quantum states is not just hard, but almost inconceivably hard. That’s great news for cryptography. www.quantamagazine.org/cryptographe...
0 replies
1 reposts
7 likes
Mathematicians recently sketched out possible scenarios for extending the geometry of space-time past a point in the distant past where the entire universe was a dimensionless dot.
www.quantamagazine.org/mathematicia...
0 replies
3 reposts
9 likes
Moiré materials, two-dimensional materials offset by a slight twist, continue to display exciting physical capabilities. This time, two distinct moiré materials have unveiled a previously unseen fundamental quantum behavior in electrons.
www.quantamagazine.org/physicists-p...
0 replies
0 reposts
3 likes
When it comes to bizarre forms of electricity, superconductivity is just the beginning. Listen to the new episode of the Quanta Science Podcast. listen.quantamagazine.org/qsp-241-s Or, read the full story. www.quantamagazine.org/meet-strange...
0 replies
0 reposts
3 likes
An interdisciplinary group of researchers have produced an equation that can predict the shape of any peat bog and thus the amount of carbon it holds.
www.quantamagazine.org/simple-equat...
0 replies
2 reposts
3 likes
Does serotonin play as big a role in depression as we previously thought? Find out on the newest episode of “The Joy of Why,” hosted by @StevenStrogatz. listen.quantamagazine.org/jow-309-s
0 replies
1 reposts
9 likes
Even though the dynamics of particle collisions are still often steeped in mystery, a table of “Before” and “After” possibilities reveals useful truths about the unchangeable rules and patterns of quantum physics. www.quantamagazine.org/the-s-matrix...
1 replies
1 reposts
4 likes
Disappointment “is both a crisis and an opportunity.” www.quantamagazine.org/how-failure-...
0 replies
0 reposts
7 likes
Decades of research showed that high frequency brain waves, known as “sharp wave ripples,” are involved in memory consolidation and storage. New research reveals how: They tag which events are worth remembering. www.quantamagazine.org/electric-rip...
0 replies
5 reposts
9 likes
Systems built with optical information processing can run more computing steps in less time and with less latency. www.quantamagazine.org/ai-needs-eno...
0 replies
0 reposts
3 likes
Alex Sushkov is supercharging an old technology for the age of the axion. He uses magnetic resonance to tip the spins of atoms. The speed that those atoms spiral back to their natural state may determine what mystery particles are lurking nearby.
www.quantamagazine.org/he-seeks-mys...
0 replies
0 reposts
5 likes
“The new algorithm is astonishingly simple and easy to implement,” said Andrew McGregor of the University of Massachusetts, Amherst. “I wouldn’t be surprised if this became the default way the [distinct elements] problem is approached in practice.” www.quantamagazine.org/computer-sci...
0 replies
4 reposts
7 likes