Twenty international research teams have been awarded a share of £2 million to use artificial intelligence in the search for new treatments for amyotrophic lateral sclerosis, the progressive neurological condition also known as motor neurone disease. Each team receives £100,000 as part of the Longitude Prize on ALS, a phased competition designed to move the most promising scientific work from early-stage discovery through to validated therapeutic targets over the next five years.

The prize does not pay out in a single round. Ten teams will be selected in 2027 to receive a further £200,000 each to build an evidence base for their proposed drug targets. By 2028, five of those will progress and receive £500,000 to pursue more rigorous validation in laboratory settings. One team, to be named in 2031, will receive a final £1 million award for producing the strongest scientific case for a viable therapeutic target.

Central to the initiative is access to what organisers describe as the largest ALS patient dataset of its kind. The resource contains genomic sequences from 9,000 patients alongside clinical and molecular data. All twenty teams will draw on this material as the foundation for their research.

The approaches vary considerably. Some teams are applying machine learning to whole genome sequencing data to identify patterns linked to disease progression. Others are mapping the gene networks that appear to protect certain neurones from degeneration, in an attempt to understand why some cells survive while others do not. One team is examining how errors in a process called splicing, which governs how genetic information is processed within cells, contribute to ALS. Another is building knowledge graphs that model how genes interact as a system rather than in isolation. Several teams are focused on identifying disease subtypes within the broader ALS population, on the basis that different biological profiles may require different treatments. One team, working in partnership with Google, will deploy the company's Co-Scientist platform to generate and test therapeutic hypotheses using the provided datasets.

The twenty teams span roughly 70 organisations across multiple countries. UK institutions involved include King's College London, the University of Oxford, University College London, the University of Sheffield, the University of Edinburgh and the University of Exeter, among others.

The MND Association is the primary funder of the prize. Tanya Curry, the association's chief executive, stated that the organisation's core mission is to achieve a world without MND, and central to this goal is providing funding for researchers to develop new treatments. She described the twenty teams as innovators whose work could deepen understanding of the condition and, in time, contribute to a cure. "MND is a devastating disease," she said, "but every step forward in research brings hope."

The prize sits within a broader shift towards using AI to interrogate large health datasets. In the United States, a National Institutes of Health-supported study has produced an algorithm trained on electronic health records to predict rare disease, with the capacity to handle incomplete or imprecise data. The system was initially tested against two rare lung conditions, where it outperformed other predictive models. In the UK, a new £500 million fund backed by the government has been launched to give early-stage AI companies access to computing infrastructure and investment of up to £20 million per startup, with a specific allocation of £282 million directed towards research and development including the creation of new datasets.

For ALS, the stakes are particular. The disease has no cure. Existing treatments can slow progression in some patients but do not stop it. Life expectancy following diagnosis is typically two to five years, though this varies. The scientific questions the Longitude Prize is trying to answer, chiefly which biological targets are most worth pursuing and why the disease manifests differently across patients, have resisted resolution for decades. Whether AI applied to a dataset of this scale can change that remains to be seen. The competition is structured on the assumption that it might.

‍