Dr Maren Westermann
Machine Learning Engineer, Max Kelsen
Dr Maren Westermann is a machine learning engineer at Max Kelsen. She works on the Immunotherapy Outcome Prediction (IOP) project that combines whole-genome sequencing and machine learning to improve the success rates of immunotherapy in cancer patients.
Maren has a strong background in the biological sciences. After completing a Bachelor’s and Master’s degree in Biology at the University of Giessen, Germany, she graduated with a PhD from The University of Queensland. In her dissertation Maren statistically analysed and modelled greenhouse gas emission from fertilised soils, and data from her PhD project can be used to improve climate models.
Maren started exploring today’s possibilities of computer science with the start of MOOCs. She started learning Python and R through online courses and integrated her programming knowledge into her research. After finishing her PhD, Maren became interested in machine learning and its potential to provide solutions for previously unsolvable problems. Building on her strength in statistics, she started educating herself in machine learning, again making use of MOOCs.
In Australia, about 3 in 10 deaths are caused by cancer making it one of the most common fatal diseases. The development of cancer has been linked to mutations of the genome. In humans, the genome comprises approximately 3.2 billion base pairs of DNA. Therefore, studying these mutations had been an insurmountable task until the turn of the millennium when the sequencing of the first whole human genome was completed. However, commercially analysing a person’s genome for detrimental mutations was still not feasible at that time because of technical and financial constraints.
Today the cost of sequencing a person’s whole genome has dropped to about 1000 USD and computers have much greater disk space and computational power, enabling researchers to handle and analyse large datasets like genomic data. Dr Maren Westermann will give an overview of state of the art machine learning models applied in today’s cancer research and highlight the challenges and constraints that are faced by the cancer research community.