A University of East Anglia-led research project has been able to use advanced mathematical modelling and machine learning called “Latent Process Decomposition” to categorise patients with osteosarcoma into different subgroups using their genetic data. Previously, all patients would be grouped together and treated using the same protocols, which has very mixed outcomes.

While genetic sequencing has previously helped to uncover different subtypes of other cancers, like breast or skin cancer, for which those patients then receive targeted treatment personalised to their cancer subtype, it has been much harder to do this with osteosarcoma – a cancer that starts in the bone and typically affects children and teenagers.