QFold: Quantum Walks and Deep Learning to Solve Protein Folding

The problem of protein folding is one of the most important and hard tasks in computational biochemistry. Recently, deep learning models, such as AlphaFold, were shown to be more effective in this task than classical techniques.

The likely advance of quantum computing could help to improve current algorithms further. A study published on arXiv.org introduces QFold, an implementation of the quantum Metropolis algorithm using the machine learning output as a starting point.

DNA helix. Image credit: Gerd Altmann Pixabay, free licence

The suggested model describes proteins according to actual torsion angles rather than using approximate rigid lattice models. A good precision could be achieved once large error-corrected quantum computers became available. Also, the time needed for calculation would shorten significantly. A proof-of-concept was successfully implemented on actual quantum hardware, thus validating the work.

We develop quantum computational tools to predict how proteins fold in 3D, one of the most important problems in current biochemical research. We explain how to combine recent deep learning advances with the well known technique of quantum walks applied to a Metropolis algorithm. The result, QFold, is a fully scalable hybrid quantum algorithm that in contrast to previous quantum approaches does not require a lattice model simplification and instead relies on the much more realistic assumption of parameterization in terms of torsion angles of the amino acids. We compare it with its classical analog for different annealing schedules and find a polynomial quantum advantage, and validate a proof-of-concept realization of the quantum Metropolis in IBMQ Casablanca quantum processor.

Research paper: Casares, P. A. M., Campos, R., and Martin-Delgado, M. A., “QFold: Quantum Walks and Deep Learning to Solve Protein Folding”, 2021, arXiv210110279. Link: https://arxiv.org/abs/2101.10279