The partnership between 1QBit, Biogen, and Accenture in 2017 explored quantum computing for pharmaceutical research and drug discovery. 1QBit, a quantum software company, brought its expertise in quantum algorithms and optimization techniques to address complex molecular modeling challenges faced by the pharmaceutical industry. Biogen, a global biotechnology company specializing in neurological diseases, provided the domain expertise and real-world drug discovery challenges. Accenture contributed its consulting capabilities and technology integration expertise to bridge the gap between quantum computing innovation and practical pharmaceutical applications.
This collaboration emerged from the recognition that traditional computational methods face significant limitations when modeling complex molecular interactions and protein folding dynamics. The partnership aimed to use quantum computing’s unique capabilities to process vast combinatorial spaces and model quantum mechanical effects that are crucial in understanding drug-molecule interactions at the atomic level.
The Drug Discovery Challenge
The primary challenge addressed by this partnership was the computational complexity of drug discovery, particularly in modeling molecular interactions and predicting drug efficacy. Traditional classical computing methods struggle with the exponential scaling of calculations required to accurately simulate quantum mechanical effects in biological systems. The computational methods used to review molecule matches and predict therapeutic effects are computationally intensive and provide limited insights into molecular structures and their interactions. These molecular comparisons are a crucial step in drug discovery where scientists analyse structural similarities between molecules to identify potential therapeutic candidates. The existing methods provided only basic similarity scores without detailed information about why molecules matched or their shared characteristics. This limits the researcher’s ability to make informed decisions about which molecular structures to pursue for further development. Additionally, the discovery process for treatments targeting complex neurological conditions is particularly challenging due to the intricacy of the central nervous system and the multifaceted nature of these diseases. These challenges result in lengthy development timelines and high costs, with neurological drug discovery often taking over a decade from concept to market, often with high failure rates in clinical trials along the way.
The challenge intensifies when dealing with complex diseases like Alzheimer’s and Multiple Sclerosis, which are focus areas for Biogen. These neurological conditions involve intricate protein misfolding and aggregation processes that are extremely difficult to model using classical computational approaches. The partnership sought to address the specific challenge of identifying novel drug candidates more efficiently by better understanding protein-drug interactions at the quantum level. Additionally, the team faced the challenge of translating theoretical quantum advantages into practical applications that could be integrated into existing pharmaceutical research workflows while working within the constraints of current quantum hardware limitations.
Solution
The quantum solution developed through this partnership focused on creating hybrid classical-quantum algorithms optimized for molecular simulation and drug-target interaction prediction. 1QBit developed specialized quantum algorithms that could run on available quantum hardware and quantum simulators to model molecular dynamics more accurately than classical methods alone. The solution incorporated variational quantum eigensolvers (VQE) for calculating molecular ground states and quantum approximate optimization algorithms (QAOA) for exploring chemical compound spaces. The team created a software platform that could seamlessly integrate quantum calculations with classical machine learning models, allowing researchers to leverage quantum advantages for specific computational bottlenecks while maintaining compatibility with existing drug discovery pipelines. The solution also included quantum-enhanced methods for protein folding prediction and molecular docking simulations, crucial for understanding how potential drug compounds interact with target proteins. By focusing on hybrid approaches, the solution could work within the limitations of current noisy intermediate-scale quantum (NISQ) devices while still providing computational advantages for specific problem instances.
Implementation
The implementation strategy involved a phased approach, beginning with proof-of-concept demonstrations on simplified molecular systems before scaling to more complex pharmaceutical targets. This included researchers at Accenture Labs collaborating with 1QBit to adapt their pre-developed structural molecular comparison algorithm and cloud-based API to include Biogen’s additional pharmacophore requirements. The team validated the quantum algorithms using quantum simulators and small-scale quantum processors to ensure accuracy and reliability. Accenture played a crucial role in designing the integration architecture that allowed Biogen’s researchers to access quantum computing resources through familiar interfaces. The process included developing APIs and software development kits (SDKs) that abstracted the complexity of quantum programming, making it accessible to pharmaceutical researchers without quantum expertise.
The team established benchmarking protocols to compare quantum-enhanced results against classical methods, ensuring that quantum advantages were measurable and meaningful. Training programs were developed to upskill Biogen’s computational biology teams on quantum computing concepts and the new tools. They also addressed practical considerations such as data security, intellectual property protection, and regulatory compliance requirements specific to pharmaceutical research. Cloud-based quantum computing resources were deployed to provide scalable access to quantum processors without requiring significant infrastructure investments. The system was designed to complement traditional molecular comparison methods, using the conventional approach to run initial comparisons on millions of molecules and then employing the quantum-enabled application to dive deeper into the most promising candidates.
Results and Business Impact
The project claimed to have yielded significant results in accelerating specific aspects of the drug discovery process. According to Govinda Bhisetti, Head of Computational Chemistry at Biogen, the solution made it “possible to rapidly pilot and deploy a quantum-enabled application that has the potential to enable us to bring medicines to people faster”. The enhanced molecular comparison capabilities allowed researchers to see exactly how and why molecular bonds matched, offering better insights to expedite drug discovery for complex neurological conditions.
Early implementations demonstrated up to 20% improvement in accuracy for certain molecular property predictions compared to classical methods alone. The quantum-enhanced algorithms showed particular promise in identifying previously overlooked drug-protein interaction sites, potentially opening new therapeutic avenues for neurological diseases. By enabling more accurate initial molecular comparisons, the solution reduced the expenses associated with screening different molecules for pharmaceutical use. This created a distinct competitive advantage through reduced time to market and cost savings. The collaboration resulted in several patent applications for quantum algorithms specific to pharmaceutical applications and established 1QBit as a leader in quantum computing for life sciences. For Biogen, the partnership provided a competitive advantage in exploring novel therapeutic approaches and reduced computational time for certain molecular simulations from weeks to days.
The business impact extended beyond immediate technical achievements, positioning all three partners as pioneers in quantum-enabled drug discovery. The collaboration attracted interest from other pharmaceutical companies and helped establish best practices for applying quantum computing in life sciences. It also contributed to the broader quantum computing ecosystem by identifying specific use cases where quantum advantages are achievable with current technology, helping to bridge the gap between quantum computing research and commercial applications.
Future Directions
The partnership continues to evolve with plans to expand the application of quantum computing to more complex pharmaceutical challenges. Future directions include developing quantum algorithms for personalized medicine applications, where patient-specific genetic data could be analyzed using quantum machine learning techniques. This includes exploring the integration of quantum computing with other emerging technologies such as AI-driven drug design and advanced imaging techniques.
As quantum hardware continues to improve, the Accenture and 1QBit aim to tackle larger molecular systems and more complex disease mechanisms. Plans include establishing a quantum computing center of excellence for life sciences applications and expanding the collaboration to include other pharmaceutical companies and research institutions. The partners are also working on developing industry standards for quantum computing in drug discovery and contributing to regulatory frameworks for quantum-enhanced pharmaceutical research. Long-term goals include achieving quantum advantage for full-scale drug discovery pipelines and potentially reducing drug development timelines by several years.
References
Accenture. (2017). “Quantum Computing in Pharmaceutical Research and Development.”
1QBit Research. (2017). “Graph-Based Molecular Similarity for Drug Discovery.”
Biogen. (2017). “Advanced Computing Applications in Neurological Drug Discovery.”
Journal of Chemical Information and Modeling. (2019). “A Quantum-Inspired Method for Three-Dimensional Ligand-Based Virtual Screening.”