It seems counterintuitive that uncertainty can help you find the most efficient way to get things done. Yet, that is the leading principle behind quantum computing. The unit of information, called a qubit (usually a particle on the quantum scale), leverages its quantum states, allowing it infinitely more computing power than the current computer bit. A handful of synchronized atoms could solve equations that would burn out the most advanced supercomputers. Governments, researchers, and military have set their sights on quantum computers for logistics problems and predictions.
Yet, the road to the next computer revolution has been slow. Qubits have been developed, but they have rarely progressed past simple math like reciting the multiplication table. The company D-wave, though, has put forward the first commercial quantum computer. It uses adiabatic quantum effects, meaning that the qubits slowly build up to a quantum energy state needed to solve the problem, instead of starting at that state to begin with. Recently, a team at Harvard, headed by Alan Aspuru-Guzik, successfully used one of D-wave’s computers to predict how proteins would fold.
Despite rising above the odds, D-wave has been at the center of controversy in both quantum mechanics and computing academia. Even while the company was rounding up big clients like Lockheed Martin and Google, may researchers didn’t believe that their product was viable, or even a quantum computer. Most notable (and colorful) are the back and forth arguments between Scott Aaronson (skeptical) and Dave Bacon (pro). Part of the problem was the difference in priorities between the scientists, who wanted the product’s quantum nature to be verified every step of the way, and the private company, which wanted to put out a workable product and build up a fan base. Still, D-wave has implemented their computer system in a few crucial benchmarks, such as using it for artificial intelligence learning at the Neural Information Processing System in 2009, and solving a complex algorithm in record time earlier this year.
Using a quantum computer to fold proteins is a big step for biology. Proteins are involved in all processes of life, and their shape determines their function. However, the different ways that each part of the protein chain can interact with another (to form the final shape) is enormous. There are twenty possible types of amino acids, the links in the protein chain, each with its own special properties. Additionally, these chains are usually 500 links long. If you take into account every potential link along each step of the chain, you can easily see how much data needs to be taken into account while studying protein folding. Also, if the proteins don’t fold by themselves perfectly, they can cause a variety of diseases and cancers. Protein folding analysis already uses classical computers, but researchers believe that quantum computing can speed up the process.
The calculations didn’t work as well as people had hoped. Out of 10,000 test runs, only thirteen yielded the right answer. However, it still worked. It’s not The Rapture, but it is definitely a step up from going through the multiplication table. As with any scientific breakthrough, it also gives valuable information to both D-wave and outside scientists about current limitations of the system, what the company needs to work on, and where they can go from here.
D-Wave computers in the lab. Photo credit: D-Wave