This project will investigate the role of circulating host and microbiome-associated metabolites in hepatobiliary cancer (HBC) development using cutting-edge multi-omics analyses. You will apply advanced Machine Learning approaches (e.g., XGBoost, Random Forest, Deep Forest, SVMs) and Mendelian Randomisation to large-scale existing metabolomics and GWAS data from major international cohorts (including EPIC, UK Biobank, and others) to identify predictive metabolic signatures for early detection, risk stratification, and prognosis.

The research combines powerful supervised learning for classification and survival modelling with causal inference methods, building on strong international collaborations. You will be co-supervised by A/Prof David Hughes (Head of MEC group, School of Biomolecular and Biomedical Science) and Dr Riccardo Rastrelli (School of Mathematics).

This fully data-driven project offers an excellent opportunity to develop expertise in ML/AI, multi-omics, and molecular cancer epidemiology while producing high-impact publications. It leverages existing datasets with strong potential for translational impact in cancer prevention and early detection.

This project will investigate the role of circulating host and microbiome-associated metabolites in hepatobiliary cancer (HBC) development using cutting-edge multi-omics analyses. You will apply advanced Machine Learning approaches (e.g., XGBoost, Random Forest, Deep Forest, SVMs) and Mendelian Randomisation to large-scale existing metabolomics and GWAS data from major international cohorts (including EPIC, UK Biobank, and others) to identify predictive metabolic signatures for early detection, risk stratification, and prognosis.

The research combines powerful supervised learning for classification and survival modelling with causal inference methods, building on strong international collaborations. You will be co-supervised by A/Prof David Hughes (Head of MEC group, School of Biomolecular and Biomedical Science) and Dr Riccardo Rastrelli (School of Mathematics).

This fully data-driven project offers an excellent opportunity to develop expertise in ML/AI, multi-omics, and molecular cancer epidemiology while producing high-impact publications. It leverages existing datasets with strong potential for translational impact in cancer prevention and early detection.