Posts Tagged ‘math’
Analytical Method Used for Stock Market Helps Pinpoint A Possible Achilles Heel for HIV
HIV's strongest sections could be its greatest weaknesses
HIV has been so difficult to fight in part because it is such an adept mutant. It produces sloppy copies of itself as it replicates, leading to many variations that can withstand drugs and vaccines. And it can produce 100 billion new virus particles every day, as Ed Yong points out over at , which leads to lots and lots of copies. Broad-spectrum drugs or vaccines can’t do very much against a target that morphs so quickly.
But not all the pieces of HIV mutate with such abandon, according to this new study. Some groups of amino acids known as HIV sectors are somewhat less fickle, staying the same while the rest of the virus morphs, according to researchers at the Ragon Institute, a research group bridging MIT, Harvard University and Massachusetts General Hospital. Researchers believe these sites must remain unchanged for the virus to survive and replicate properly.
Researchers led by HIV research pioneer Bruce Walker and MIT chemical engineering professor Arup Chakraborty say this stalwart section of the virus can be turned against it. If the immune system can be trained to attack all the amino acid portions in an HIV sector, the virus will either have to mutate to thwart the attack — thereby undermining its structural integrity, crippling itself — or not mutate, which would render it helpless against drugs or vaccines.
This new targeted approach came from Chakraborty, who thought Walker and colleagues were too limited in their search for solutions, Yong reports. The team turned to random matrix theory, developed in the 1950s to solve nuclear physics problems and which has been used to analyze stocks, as the notes. It can pinpoint correlations between groups of objects, so it can assess how one stock is linked to other groups of stocks, for instance.
Working with HIV proteins taken from a massive database, the team used random matrix theory to analyze HIV’s genetic code and find groups of amino acids whose mutations were coordinated. The segment that mutated the least was dubbed sector 3, on an HIV sector known as Gag, which makes up HIV’s honeycombed inner shell. If the shell mutates, the honeycomb won’t lock together, and the virus would collapse.
“Multiple mutations within this sector are very rare, indicating previously unrecognized multidimensional constraints on HIV evolution,” the authors write in a on their research, which is published this week in the Proceedings of the National Academy of Sciences.
Incidentally, a rare group of patients who can fight HIV without drugs — known as “elite controllers” — use their own immune systems to attack sector 3.
All this suggests a new way of thinking about HIV treatment, the WSJ and others point out. Perhaps HIV drugs should dispense with the full-on assault and opt for targeted strikes instead.
Buoyed by this research, other teams are reportedly already planning new animal studies to test just that.
[via , ]
Largest DNA-Based Computer Ever Built Can Calculate Square Roots
Researchers at Caltech engineered the most complex biochemical circuit ever created from scratch, according to a new paper published today. The circuit uses DNA instead of electronic transistors to produce the on-off, and-or signals that allow a computer to conduct its calculations.
In a typical computer, transistors let a current of electrons flow in and out. The DNA computer instead uses pieces of short, single-stranded DNA or partially double-stranded DNA placed in a test tube of salt water. The strands stick out like tentacles from the DNA’s double helix, as a news release from Caltech . The DNA molecules collide in the water and bind together, producing and releasing offspring molecules. These act as the signals, like electrons in a traditional chip, and they travel among the DNA “gates,” connecting the circuit.
Pairs of gates can create and-or logic based on the output molecules observed, as Ars Technica explains it. (Check out for a more thorough explanation of how this works.)
The researchers, led by postdoctoral researcher Lulu Qian, can encode whatever DNA sequence they want, so they have full control over how the DNA strands interact.
Their largest computer was a 74-molecule, four-bit circuit that could compute the square root of any number up to 15, rounding down the answer to the nearest whole number. To get the answer, the researchers would monitor the concentration of output molecules in the test tube, using fluorescent tagging.
The process takes a long time, but speed is not the point — using this method, scientists could eventually engineer biochemical pathways that are capable of making decisions. This type of control over chemical reactions could be useful for anything from pharmaceuticals to industrial processes. Imagine DNA-based computer chips embedded in your skin, releasing drugs when the time is right, or a DNA computer that can study the concentration of certain molecules in a blood sample and quickly diagnose a disease.
The circuit can be scaled up to larger DNA computers, the researchers say. They can also be customized by adjusting the types of DNA used or reconfiguring the circuit.
“We want to make better and better biochemical circuits that can do more sophisticated tasks, driving molecular devices to act on their environment,” Qian said in a news release. The computer was described in a paper in today's issue of the journal Science.
[, ]
Science Fiction Writer Envisions How Smart Algorithms Could Streamline the US Tax Code
Brin's proposal aims to replace the document President Obama once referred to as a "10,000-page monstrosity" with a mathematical model that can find the simplest "no losers" code. The idea is to create software that can input a round of simplifications and then output a spreadsheet of how those simplifications would affect, say, a hundred different groups, from small business owners to students to corporations. You'd give it some boundaries to ensure that none of those groups take a hit of more than, say, 5% (which sounds kind of high to me, but it's the idea that counts), and keep feeding it simplifications until you arrive at the simplest code that doesn't screw anyone.
This isn't a "how to fix the tax code" or "how to block loopholes" or "how to make Goldman Sachs ." It's just designed to keep the bottom line of taxes as close to what they are now as possible while eliminating as much verbiage, repetition, and unnecessary language as possible. It might require a balancing act--if one round of simplifications screws one group, the next round might have to balance that out.
Brin isn't convinced anyone will actually attempt this, though he says "some people who know about such things" say it's not only possible but already exists in bits and pieces--various government agencies use programs like this to test specific inputs and outputs. But it certainly seems like something we should try.
[ via ]
Improve Your Math Skills With An Electric Jolt To Your Brain*
Patients were asked to learn new symbols to represent numbers, then, while they were on TDCS, they attempted to organize the numbers. Participants whose brains were being stimulated demonstrated an improved ability to perform the task. The amazing part is that, when tested again six months later, they retained their higher performance level. The current helps the affected nerves to fire more quickly, making it easier to learn information.
The next trials will involve patients who have lower-than-average number processing skills, and Oxford scientists hope to one day develop a device to deliver TDCS. While it may be some time before such brain-zapping is widely administered, this treatment could help the significant portion of the population (nearly 20 percent) with moderate to severe math disability, and possibly those with difficulty in other subjects as well.
* Do not zap your brain with electricity except under professional supervision.
Improve Your Math Skills With An Electric Jolt To Your Brain*
Patients were asked to learn new symbols to represent numbers, then, while they were on TDCS, they attempted to organize the numbers. Participants whose brains were being stimulated demonstrated an improved ability to perform the task. The amazing part is that, when tested again six months later, they retained their higher performance level. The current helps the affected nerves to fire more quickly, making it easier to learn information.
The next trials will involve patients who have lower-than-average number processing skills, and Oxford scientists hope to one day develop a device to deliver TDCS. While it may be some time before such brain-zapping is widely administered, this treatment could help the significant portion of the population (nearly 20 percent) with moderate to severe math disability, and possibly those with difficulty in other subjects as well.
* Do not zap your brain with electricity except under professional supervision.
Bees Solve Hard Computing Problems Faster Than Supercomputers
Yet another reason to save them from extinction
Bumblebees can solve the classic "traveling salesman" problem, which keeps supercomputers busy for days. They learn to fly the shortest possible route between flowers even if they find the flowers in a different order, according to a new British study.
The traveling salesman problem is an http://en.wikipedia.org/wiki/NP-hardNP-hard (read: very hard) problem in computer science; it involves finding the shortest possible route between cities, visiting each city only once. Bees are the first animals to figure this out, according to Queen Mary University of London researchers.
Bees need lots of energy to fly, so they seek the most efficient route among networks of hundreds of flowers. They navigate using angles of sunlight, which helps them find their way home, researchers say. To do this, their tiny brains must pack a powerful memory.(, according to a separate study that came out last week.)
To test bee problem-solving, researchers Lars Chittka and Mathieu Lihoreau tested bees’ response to computer-controlled artificial flowers. They wanted to see whether the bees would go after the flowers in the order in which they were discovered, or if they would figure out the shortest route among all the flowers even as new ones were added. The bees explored the locations of the flowers and quickly figured out the shortest paths among them, according to a Queen Mary news release.
This is no small feat, especially considering bee brains are about as big as a microdot. When it comes to intelligence, size apparently does not matter.
Earlier this year, researchers showed that bees because they can make out the relative patterns that make up a face. The new research further suggests bees are highly sophisticated problem solvers, and that better understanding of their brains could improve our understanding of network problems like traffic flows, supply chains and epidemiology.
The research will be published this week in the journal The American Naturalist.
[]
Celebrating Benoit Mandelbrot, the Man Who Made Math Beautiful
Mandelbrot wrote mathematical formulas that help explain nature, supplementing the cold, sharp angles of basic Euclidian geometry with fantastic spirals, asymmetric tendrils and repeating bubbles. With his formulas, complex structures like coastlines could be explained with a little neat math.
In the intro to his book “The Fractal Geometry of Nature,” he asks, “Why is geometry often described as cold and dry? One reason lies in its inability to describe the shape of a cloud, a mountain or a tree.”
In fractal shapes — which Mandelbrot coined from the Latin word fractus, or “broken” — each part mimics the pattern of the whole. Magnifying each part reveals ever more complexity, repeating in an infinite cycle.
The Mandelbrot set is basically a set of complex numbers, which belong to one side of an equation or not. Images can be made by assigning colors to each number. Mandelbrot completed his first fractal visualization at IBM’s Thomas J. Watson Research Center in New York in 1980. His work was perfectly suited to the nascent world of computers, but it helped us understand natural phenomena better than ever.
Mandelbrot showed that very simple formulas can yield extraordinarily complex results. Fractals can be used to model everything from broccoli heads and mammalian brains to stock markets and the distribution of galaxies.
of some stunning fractal images in honor of the late Mandelbrot. Click the thumbnails to launch the gallery.