Moderna is collaborating with IBM to develop a quantum-enabled biotechnology pipeline, aiming to augment—not replace—classical computing for mRNA medicine design. Researchers have applied Conditional Value at Risk (CVaR), a risk-assessment technique from finance, to variational quantum algorithms (VQAs) to improve performance and identify optimal mRNA sequences, achieving results matching commercial classical solvers. In 2024, the team published research demonstrating a VQA execution using up to 80 qubits and mRNA sequences up to 60 nucleotides—the largest sequences simulated on a quantum computer to date—with further research planned for 2025 expanding to 156 qubits and 950 non-local gates.
mRNA and Quantum Computing
Moderna is collaborating with IBM to develop a quantum-enabled biotechnology pipeline, seeking to scale progress in mRNA medicine design rather than awaiting full quantum technology maturity. A key challenge lies in optimising the vast number of possible mRNA sequences to accurately instruct the body to produce therapeutic proteins, a task where quantum computing offers a complementary approach to classical methods. While classical computers are powerful, they encounter limitations with computationally intensive problems, particularly in identifying stable and effective nucleotide sequences that avoid unwanted immune responses.
Understanding mRNA behaviour requires predicting its secondary structure—the pattern of internal attraction between nucleotides that influences protein translation and stability. Predicting this structure presents a complex combinatorial optimisation problem, ideally suited for quantum-enhanced algorithms due to the astronomically large number of theoretical folding possibilities. Researchers are exploring variational quantum algorithms (VQAs) – a class of algorithms for near-term quantum applications – with the potential to deliver quantum advantage before the advent of fully error-corrected quantum computers.
To improve VQA performance, Moderna and IBM researchers have applied Conditional Value at Risk (CVaR), a risk-assessment technique from finance, to focus the optimisation process on the most promising solutions. CVaR concentrates on the lower tail of the energy distribution, effectively mitigating variance and steering the classical optimiser towards optimal solutions without substantial computational overhead. This approach is advantageous as it reduces the burden of suppressing hardware noise and the need for additional error mitigation techniques, allowing for more efficient use of both quantum and classical resources.
The joint Moderna-IBM Research team has demonstrated a quantum approach matching the performance of commercial classical solvers for combinatorial optimisation problems, publishing research in the IEEE International Conference on Quantum Computing and Engineering in 2024. This research applied CVaR-based VQAs to mRNA secondary structure prediction, achieving one of the largest and most advanced VQA executions to date, involving up to 80 qubits and mRNA sequence lengths up to 60 nucleotides. Further research, scheduled for publication in 2025, expands this to 156 qubits and 950 non-local gates, alongside a new approach called instantaneous quantum polynomial (IQP) circuit-based quantum optimisation.
Moderna’s goal is to build a near-term quantum-enabled biotechnology pipeline, augmenting classical methods with quantum computing to generate a wider range of molecules for testing. IBM similarly envisions a collaborative approach, dividing tasks between classical and quantum architectures to rapidly deliver results, focusing on quantum-centric approaches to the secondary structure problem at increasingly larger scales. Moderna values IBM’s track record of research delivery and its clear roadmap for quantum technology development, aiming to leverage quantum computing to maximise the impact of mRNA medicines as the technology scales.
Optimising mRNA Sequence Design
In 2024, the Moderna-IBM Research team published research in the IEEE International Conference on Quantum Computing and Engineering, applying CVaR-based VQAs to the mRNA secondary structure prediction problem, achieving one of the largest and most advanced VQA executions to date, involving up to 80 qubits and mRNA sequence lengths up to 60 nucleotides—the largest sequences simulated on a quantum computer to date. Further research, scheduled for publication in 2025, expands this to 156 qubits and 950 non-local gates, and introduces a new approach called instantaneous quantum polynomial (IQP) circuit-based quantum optimisation.
The collaboration is focused on quantum-centric approaches to the secondary structure problem at even larger scales, with the ultimate aim of building a near-term quantum-enabled biotechnology pipeline, rather than replacing classical computing altogether. This pipeline is envisioned to generate a wider range of molecules for testing, augmenting classical methods with quantum computing to maximise the impact of mRNA medicines.
Moderna values IBM’s track record of delivering research results and its clear roadmap for developing quantum technology, anticipating that as quantum computing scales, it will be leveraged to further optimise mRNA medicine development.
Variational Quantum Algorithms and CVaR
CVaR-based VQAs reduce the burden of suppressing hardware noise, allowing for more efficient use of quantum and classical resources. As quantum computers scale and become more robust, this approach is expected to deliver reliable results for problems currently intractable with classical methods. IBM anticipates seeing examples of quantum advantage by 2026, contingent on collaboration between the quantum and high-performance computing communities.
The joint Moderna-IBM Research team has achieved impressive results, demonstrating a quantum approach that matches the performance of commercial classical solvers for combinatorial optimisation problems. In 2024, they published research in the IEEE International Conference on Quantum Computing and Engineering, applying CVaR-based VQAs to the mRNA secondary structure prediction problem, achieving one of the largest and most advanced VQA executions to date. This involved up to 80 qubits and mRNA sequence lengths up to 60 nucleotides—the largest sequences simulated on a quantum computer to date. Further research, to be published in 2025, expands this to 156 qubits and 950 non-local gates. They also presented a new approach, called instantaneous quantum polynomial (IQP) circuit-based quantum optimisation.
Moderna’s ultimate goal is not to replace classical computing, but to build a near-term quantum-enabled biotechnology pipeline. Quantum computing can offer a more diverse set of solutions, generating a wider range of molecules for testing. This augmented approach, combining classical and quantum methods, is envisioned as the most realistic scenario. IBM similarly envisions classical and quantum methods working together to solve complex problems, dividing tasks between architectures to rapidly deliver results. The team is focused on quantum-centric approaches to the secondary structure problem at even larger scales. Moderna values IBM’s track record of delivering research results and its clear roadmap for developing quantum technology. As quantum computing scales, Moderna aims to leverage it to maximise the impact of mRNA medicines.
A Hybrid Classical-Quantum Pipeline
The collaboration seeks to establish a near-term quantum-enabled biotechnology pipeline, recognising that quantum computing can offer a more diverse set of solutions, generating a wider range of molecules for testing. This augmented approach, combining classical and quantum methods, is envisioned as the most realistic scenario, with IBM similarly envisioning classical and quantum methods working together to solve complex problems by dividing tasks between architectures to rapidly deliver results.
The team remains focused on quantum-centric approaches to the secondary structure problem at even larger scales, with Moderna valuing IBM’s track record of delivering research results and its clear roadmap for developing quantum technology.
Further research, scheduled for publication in 2025, expands the current work to 156 qubits and 950 non-local gates, alongside the presentation of a new approach called instantaneous quantum polynomial (IQP) circuit-based quantum optimisation. This builds upon the 2024 publication in the IEEE International Conference on Quantum Computing and Engineering, which detailed the application of CVaR-based VQAs to the mRNA secondary structure prediction problem, achieving one of the largest and most advanced VQA executions to date involving up to 80 qubits and mRNA sequence lengths up to 60 nucleotides—the largest sequences simulated on a quantum computer to date.
More information
External Link: Click Here For More
