Quantum computing is perhaps Silicon Valley’s biggest buzzword. After Google and NASA jumped on the bandwagon several years ago — first with D-Wave Systems, then striking out on their own — the excitement bubbled over to longstanding corporations and fledgling companies alike. Now, new startups like Rigetti Computing and longtime competitors like Microsoft have joined the race to build the most powerful computers in history.

Quantum computers, however, wouldn’t just belong in the basement of Google’s headquarters. Potential applications show up in some very surprising places — perhaps most importantly of all, in drug production.

A theoretical quantum computer would use a “qubit” — a quantum bit — as an analog to the bits in your normal, run-of-the-mill computer (called a classical computer). A bit in a classical computer has two possible states — 0 or 1 — and all higher-level methods of transferring information are based on combinations of these states. Qubits, however, can exist in a so-called superposition of states. While a classical bit is limited to either a 0 or a 1, a qubit can be a combination of both states.* Computationally, this represents a significant advantage, and computer scientists have demonstrated that quantum computers could theoretically outspeed classical computers on certain types of problems.

But why? What’s the point? Despite Moore’s Law and our eternal desire for the latest MacBook Pro, the average user will never require the type of speedup that researchers look for. While quantum computing has applications in solving many difficult problems, among the most remarkable is drug synthesis. Simulating drugs — for example, how a cancer drug interacts with a tumor and other body cells — is computationally a very difficult problem. Current methods in drug synthesis involve significant approximations on the molecular and atomic level. While a more precise approach would be preferable, the time required to accurately simulate drugs without such approximations is prohibitive.

Quantum computation, however, is ideally suited to the task. On a molecular level, interactions are quantum-mechanic. The processes that govern chemical interactions are fundamentally linked to those that would allow for the existence of a quantum computer. This, at least in theory, would make a universal quantum computer an ideal approach to drug discovery, and could potentially result in both faster drug production and more precise drugs. Forms of cancer that were previously untreatable could be more effectively targeted and destroyed, with fewer malignant side effects on the body. The pharmaceutical benefits are substantial, because quantum computation essentially eliminates the need for guesswork in building up a model of the human body and how it reacts to stimuli.

The processes that govern chemical interactions are fundamentally linked to those that would allow for the existence of a quantum computer... a universal quantum computer would be an ideal approach to drug discovery.

Clearly, quantum computers are still years away, and quantum computation is still a very active field of research. Qubits suffer from a Sisyphus-level problem — they need to preserve the state of the system after performing an operation, and we need to retrieve that data without changing the state of a system. Researchers call the ability of a quantum system to maintain its state coherence, and it’s easily one of the largest headaches in the field today. To maintain coherence and prevent external noise — such as heat — from affecting a system, qubits currently require massive fridges to stay at a few degrees above absolute zero. The resources required to keep qubits running is industry-scale.

While quantum computing still has years to go before its applications are realized, steps in that direction have already begun. This past July, a collaboration between Harvard, Lawrence National Labs, UC Santa Barbara, Tufts University, University College London, and Google produced a quantum simulation of a single molecule, solving something called the molecular electronic structure problem, which is essential for predicting chemical reaction mechanisms. While quantum simulations have been attempted before, they’ve always required expensive pre-processing steps on a classical machine. As quantum simulation on scalable hardware becomes more viable, we could potentially predict precise values for chemical reaction rates and design new pharmaceuticals and catalysts. Even with relatively simple molecules, classical computers struggle with this type of simulation, because the memory required grows exponentially. Quantum computers, however, have no such problem, and the result as released by Google points towards a very exciting future.

* If this feels confusing to you, you’re in good company. The following quote is often attributed to Richard Feynman: “If you think you understand quantum mechanics, you don’t understand quantum mechanics.”

About The Author

Rohit Dilip