Quantum Machine Learning for Genomics: Assessing Real-World Potential and Constraints

Sequencing machines now read trillions of base pairs of DNA, but the algorithms that make sense of that data can only handle simple, pairwise relationships between genes. This project asks whether quantum machine learning can break that bottleneck. The core problem is that biological systems depend on complex, higher-order interactions among genes, cells, and environmental factors—yet most current analyses are limited to associations between just two variables at a time. Modelling those richer correlations with classical computers is computationally expensive, slow, and energy-intensive. If quantum machine learning proves practical for genomics, it could mean faster training of models, better generalisation to new data, and lower energy consumption for large-scale analyses. That could accelerate how researchers translate raw genetic sequences into an understanding of disease mechanisms. The project is not building a working quantum computer; it is testing whether today’s quantum technology can handle simplified biological datasets at all, and what hardware resources would be needed for a real advantage. In collaboration with the UK National Quantum Computing Centre, the team will construct test models, identify suitable datasets, and run community-building activities to guide future partnerships between quantum computing and genomic science. The work is exploratory and fundamental—it assesses potential and constraints, not a guaranteed breakthrough.