Applications of artificial intelligence for leveraging genetic data to inform drug development

Summary

The recent explosion of genetic data availability offers unprecedented opportunities to revolutionize drug discovery and development.

Project aims

This PhD aims to explore a spectrum of artificial intelligence (AI) techniques to harness genetic information for identifying and validating novel drug targets.

By integrating methods such as clustering algorithms, graph neural networks, and natural language processing (NLP) including large language models, the research seeks to uncover hidden patterns and insights within complex genetic datasets.

Novel clustering algorithms specifically targeted at phenotypic and genetic data will be employed to detect patterns and pathways, potentially revealing underlying biological pathways and causal relationships relevant to disease mechanisms.

NLP and advanced language models will be utilized to extract and synthesize information from vast biomedical literature and databases, streamlining processes such as instrument selection in genetic studies.

The interdisciplinary approach bridges biology, clinical knowledge, and data science, aiming to develop innovative strategies for interpreting genetic information.

Techniques such as Mendelian Randomization will be leveraged to establish causal relationships, while AI models will assist in predicting drug-target interactions and potential side effects.

By adopting a broad range of AI methodologies, the PhD aims to enhance the efficiency and accuracy of drug target discovery and validation. The ultimate goal is to contribute to translating genetic insights into actionable therapeutic interventions that can improve patient outcomes.

Contact details

Stephen Burgess - sb452@medschl.cam.ac.uk

Opportunities

This project is open to applicants who want to do a:

• PhD