Identification of biomarkers that predict pancreatic cancer response to targeted drug therapy

Project aiming to identify biomarkers by implementing a machine learning approach based on the use of genomic, transcriptomic and drug response data generated from organoids derived from pancreatic cancer patients.

Pancreatic cancer is predicted to become the second leading cause of cancer-related death within the next 10-15 years. In contrast with other cancers, very little improvement has been made in clinical outcomes, and very few treatments are available for patients with this disease.

Consequently, the 5-year survival rate for pancreatic cancer patients is still lower than 10%. Recent advances in molecular characterization have led to the identification of multiple genetic alterations that contribute to driving this disease, raising the hope that therapeutics designed to target these alterations may significantly improve patient outcomes. Yet, biomarkers that predict the response of individual tumours to a given drug remain elusive.

In this project, we propose to identify such biomarkers by implementing a machine learning approach based on the use of genomic, transcriptomic and drug response data generated from organoids derived from pancreatic cancer patients.

The aim will be to develop clinically meaningful markers that allow matching of individual patients with drugs that are the most likely to prove effective in treating their tumour. The project provides an exciting opportunity for an excellent student willing to combine a small amount of laboratory work (organoids drug assays) with computational biology/machine learning implementation.

The project would suit a candidate with experience in laboratory techniques who is keen to develop skills in computational biology. It is expected that the student would have some previous familiarity with one or more programming languages and a keen ability to learn more computational techniques.

The project may also be suitable for candidates who have undertaken bioinformatics course who wish to develop expertise in applied machine learning.

Contact and more information

Primary supervisor

Professor Frédéric Hollande
frederic.hollande@unimelb.edu.au

Co-supervisor

Professor Lachlan Coin
lachlan.coin@unimelb.edu.au

Tumour Heterogeneity in Metastatic Cancer
University of Melbourne Centre for Cancer Research and Department of Clinical Pathology

Level 10, Victorian Comprehensive Cancer Centre