Reimagining drug discovery with quantum computing

March 17, 2021

The discovery and design of novel therapies is an increasingly challenging endeavour in the 21st century; though technology has progressed in leaps and bounds, identifying previously unknown molecules and drugs have become trickier, and every avenue that could accelerate or enhance the process must be explored. Approaches that were once unthinkable, such as artificial intelligence, have become commonplace in drug discovery – and now, quantum computing is emerging as pharma’s next frontier.

Unlike classical computers, which rely on bits that are either on or off, quantum computers use qubits, which can either be on or off, or both – known as the superposition. This superposition allows quantum computers to make multiple calculations at one time, and with far greater efficiency than traditional technology allows.

 

The advantages of quantum computing

Though the application of quantum mechanics in pharma is still in its infancy, quantum computers’ capacity to dramatically accelerate and optimise testing and predictions makes it a natural partner for the drug discovery space – and the life sciences are keen to get involved.
Quantum computers have a natural advantage over classical systems when it comes to simulating molecules for in silico drug discovery and design.
Quantum computers behave by the same laws as the molecules themselves, which means that you can be much more efficient at simulating a molecule using a quantum computer than you can with a classical computer when it’s about simulating two atoms, then for a quantum computer, it only needs two qubits to do that simulation. But for a classical computer, it would need at least four bits to simulate.

This is an example of how the resources for a classical computer explode to simulate something quantum mechanical like a molecule, compared to a quantum computer.

 

A different approach

Quantum computing is diamond-based computers that are a fraction of the size of typical their mainframe and can run without the extreme conditions usually required. Because the hardware differs from that of traditional mainframes, it can be used differently.

There are two approaches to computational chemistry: simulating a single, complex molecule to understand its properties, or studying the interactions within a large system of molecules. Most quantum computing hardware tackles the former – but the latter, known as molecular dynamics, is where it’s focused.

To understand that reaction, and all of the complexity involved in that reaction around what is produced, what inhibits the production of things that we want, what the by-products are, and how it can stimulate and drive that chemical reaction.

The reactions simulated in molecular dynamics are the same as those seen in enzymes, which are widely used as the basis for many therapeutic products.

This application of molecular dynamics is already pursued by classical computing technologies, Quantum Companies have invested huge amounts of effort and money into building molecular dynamic engines using classical computers, and essentially seeking to replace these with our quantum-accelerated molecular dynamics engines.”

The vision is to build a “massively parallelized” system in which quantum accelerators and classical computers are deeply integrated, and therefore able to simulate several complex molecules and their reactions at once.

 

Work in progress

The accelerators could revolutionize in silico drug discovery, but the technology itself is still under development.

It’s challenging right now because the hardware is evolving, it’s not quite at a scale where it can be used today to solve the tough, important problems that people are trying to tackle.
There is some time yet before hardware gets to the point of sufficient performance and sufficient scaling, being able to encode enough information to solve the problems.”

But that doesn’t mean people shouldn’t be thinking about implementing quantum accelerators in their drug research projects. The company anticipates its devices will be poised for application in pharma in the not-so-distant future – within three to five years, it hopes – and we have to be optimistic about what the technology would mean for drug research.

“[Quantum accelerators] dramatically lower the barrier to doing computational simulations that accelerate drug design, and the understanding of chemical processes that you’re observing in the lab.

It’s about really lowering that barrier to having adequate computational resources, to using computational chemistry to help to design and accelerate the research by understanding what experiments are showing.