Using Knowledge Graphs to repurpose drugs for Motor Neurone Disease and Frontotemporal Dementia

There are no effective drugs to significantly modify the course of two serious medical conditions: motor neurone disease (MND) and frontotemporal dementia (FTD). These conditions are devastating: average time to death after diagnosis for MND is 2-3-years, and for FTD is 5-7-years. Clinical trials to test potential drug candidates have met with failure.

My aim is to use data science techniques to identify pre-existing drugs which can be repurposed to treat MND and FTD. This approach can produce treatments more quickly and cheaply. I will use a knowledge graph, which represents highly complex biological relationships between different data types.

This includes data about the building blocks of life - genes, gene expression and proteins, as well as drugs and drug targets. It will also include information from brain scans.

I will use data science techniques to analyse the knowledge graph, identifying existing drugs that could be repurposed for MND or FTD treatment. This work will have two outputs. First, drug candidates for MND and FTD, which can be put into clinical trials.

Second, the development of a methodological pipeline to identify new treatments, which could be applied across other neurological diseases e.g. Alzheimer’s disease or multiple sclerosis.