Researchers at the Cleveland Clinic and IBM are laying the groundwork for applying quantum computing methods to protein structure prediction.
Their ultimate goal? To “design quantum algorithms that can find how to predict protein structures as realistically as possible,” according to IBM researcher Dr. Hakan Doga, who spearheaded the research team with Cleveland Clinic postdoctoral fellow Dr. Bryan Raubenolt.
The function of protein molecules in cells depends on more than just the amino acids that make up their chains. Protein structure – its folding – also determines how a protein functions and binds to other molecules in the body, which in turn determines many aspects of human health and disease.
Accurately predicting protein folding would allow researchers to better understand how diseases spread and, ultimately, how to develop effective therapies.
In recent years, scientists have tried to predict protein structure using two methods. The first method, which relies on machine learning, is limited by the number of proteins the machine has been taught to recognize. The result could be inaccurate when the machine encounters a protein that is mutated or very different from those on which it was trained. That’s the sort of problem that can arise with genetic disorders.
The second method simulates the physics of protein folding. Simulations allow researchers to look at a given protein’s various possible shapes and find the most stable one. The most stable shape is critical for drug design.
But beyond a certain protein size, simulations are nearly impossible on a classical computer. That’s where quantum computing comes to the rescue.
The IBM/Cleveland Clinic team applied a mix of quantum and classical computing methods to their simulations. The mix allowed quantum algorithms to address the areas that are challenging for classical computing, including protein size, intrinsic disorder, mutations, and the physics involved in protein folding.
The effectiveness of the mix was validated by accurately predicting the folding of a small fragment of a Zika virus protein on a quantum computer, compared to state-of-the-art classical methods. The quantum-classical hybrid framework’s initial results outperformed both a classical physics-based method and AlphaFold2.
“This work is an important step forward in exploring where quantum computing capabilities could show strengths in protein structure prediction,” commented Dr. Doga.
The study was published the Journal of Chemical Theory and Computation.