Learning the next generation of drug targets by modelling diseases, targets and their relationships

Target selection is the first and arguably most important decision in any drug discovery programme. Drug development is a lengthy, costly process with remarkably low success rates, and a substantial amount of failures is due to poor target selection strategies. In this proposed project we aim to tackle this issue by leveraging the wealth of biomedical data available in the public domain and state-of-the art data mining and machine learning methods. The key deliverables of the project will be (1) identification of features defining successful drug targets; (2) discovery of novel candidate targets; (3) publication of results. The project will be aligned to the "Biological informatics" and the "Artificial intelligence technologies" research areas of the EPSRC.