Current drug development processes require on average 10-15 years to provide a new medicine, leaving a long wait for patients to experience clinical benefit. Recent developments in artificial intelligence (AI) are enabling processes to be accelerated, and this research proposal aims to confirm whether AI can speed up particular steps in the drug development process. In particular, AI can support process improvements in the pre-clinical stages of drug development, i.e., before medicines are used in humans. For example, aim (i) of this study is to use to AI to understand the biological links between genes and diseases, by processing large volumes of data rapidly. In addition, in aim (ii) we plan to use AI models to enable accelerated design of diagnostic tests, to understand if a drug is fit for purpose (biomarker tests). Previous work we have undertaken has demonstrated AI models can be useful in drug development – for example, in predicting the structure of proteins, which are the key targets that medicines act against. Traditional methods to understand protein structure can take years of experimental work in a laboratory, while AI models can make immediate predictions, hence allowing drug design work to progress more quickly. This study will use genetic, imaging and blood test data, compared with health outcomes, to develop AI models to support drug development efforts. The study will run over three years and results from this work will support the development of new medicines for chronic diseases.
