Project Lead: Curenetics
Sector: Healthcare
Project Summary
The project addresses a key challenge in personalised cancer vaccines and immunotherapies: accurately predicting which peptides will be processed by tumour cells, presented on human leukocyte antigen (HLA) molecules, and recognised by the immune system. These predictions involve highly complex, non-linear biological interactions and large, diverse datasets that are difficult for conventional machine-learning approaches to model effectively.
The team will develop and benchmark a hybrid quantum-classical machine-learning approach for antigen-processing and pan-allele HLA-binding prediction. Classical machine-learning baselines will be established first, followed by the integration of quantum feature representations and Variational Quantum Circuits (VQCs). Testing will be carried out using simulators and IBM Quantum physical QPUs.
Expected outcomes include benchmarked hybrid prediction models, a detailed assessment of quantum hardware requirements and constraints, comparison of classical versus quantum-hybrid performance, and an evidence-based roadmap for applying quantum computing to cancer vaccine and immunotherapy discovery.
