Principal Investigator: Professor Francesca Buffa
Institution: University of OxfordTags: 43313, cancer, GWAS, hypoxia, Machine Learning, Metabolism, physiology
Low oxygen levels (hypoxia) is a feature of solid tumours and is associated with changes within the tumour. Hypoxia is related to several factors in the patient, including the structure of their tumour, the way the tumour uses energy (metabolism) and other host features (e.g. blood pressure). We have published extensively on hypoxia and metabolism and want to investigate how pathways associated with these linked areas relate to clinical characteristics, medical conditions and your genetic code (DNA).
Low oxygen levels (hypoxia) in solid tumours is associated with poor patient outcome, partly by changing the way cells behave but also by limiting the effectiveness of cancer treatments. Therefore, over the last few decades, hypoxia and its associated pathways have become important targets for new therapies. To date, we have focussed on identifying the key pathways associated with hypoxia by looking at the expression of genes within tumours. However, little is known about these pathways in patients with and without cancer. It is possible that mutations in some of these pathways may predispose people to cancer and we want to investigate this. To date, there has been no approved research related to hypoxia using the UK Biobank data.
We will use a number of advanced bioinformatics and machine learning approaches.
We shall use the genetic and clinical information within the Biobank to see if different patterns within genes related to hypoxia-associated pathways may influence clinical outcome, e.g. in this case the risk of developing cancer. We will use various sophisticated statistical techniques to explore these associations. As part of exploring the clinical measures/features within the biobank, we will use various methods to identify the most informative attributes related to hypoxia and its associated pathways, including machine learning. This will help inform further studies in the future. By looking at the genetic and clinical information, we will have two different viewpoints of hypoxia-associated pathways within patients. The UK Biobank provides a good platform to look at hypoxia-associated pathways as it is the largest prospective study to date that includes genetic data.
Public health impact
Our work will highlight genes within hypoxia-associated and metabolic pathways which might be associated with cancer development. This potentially may influence cancer screening, drug discovery and treatment selection in patients. Furthermore, comparing clinical measurements will help inform future studies.