A University of Dundee researcher has been awarded £1.6 million in funding from UK Research and Innovation (UKRI) to develop machine learning tools for the life sciences.
Machine learning has rapidly transformed many aspects of our daily lives in the last decade. It is also increasingly playing a vital role in analysing biomedical data. However, using these techniques to answer fundamental, mechanistic questions about biological processes remains challenging.
Dr Gabriele Schweikert, a researcher at the University’s School of Life Sciences, uses machine learning tools to better understand important molecular processes in living cells and how these go wrong in diseases. She has been awarded a UKRI Future Leaders Fellowship to build a team in the Division of Computational Biology with the required interdisciplinary competence to maximise the potential of machine learning in scientific research.
“Currently few people are equipped to tackle the challenges that arise in the interface between advanced computational biology and wet-lab biology,” said Dr Schweikert.
"The rapid progress in machine learning technology as well as in experimental high throughput measurements, make it possible to find hidden patterns in large, complex data sets. These patterns can then be used to make predictions in similar circumstances that have not yet been observed.
“However, many questions that motivate researchers in the biomedical sciences are about the underlying causes of the observations. While algorithmic tools are available to tackle these questions as well, they are not widely used in practice. In most instances, causes cannot be computed from data alone, as they require additional knowledge of the data-generating process.”
Dr Schweikert initially trained as a physicist, but became interested in the biological sciences early in her undergraduate studies. During an interdisciplinary PhD at the Max Planck campus in Germany, she developed new computational systems for DNA sequence analysis. She started her own research group on computational epigenomics at Dundee last year and also holds an affiliation with the Cyber Valley Initiative, Germany.
Dr Schweikert will work with Professor Tom Owen-Hughes, Head of the Gene Regulation and Expression Unit within the School of Life Sciences. They will design and conduct experiments and to develop specific data analysis tools to better understand epigenetic regulation of gene expression.
Epigenetic patterns are chemical modifications on top of the DNA that do not change the genetic code. Instead, they provide an intricate system of bookmarks that allow cells to control and remember which genetic programs are active in a given cell. They package and organise the DNA, such that certain parts are stored away while others are readily executable.
Epigenetic mechanisms are vital for our health and epigenetic malfunctioning has been observed in human diseases. A more in-depth understanding of epigenetic patterns will contribute to new insights of the epigenomic factors underlying diseases, including cancer and neurological and autoimmune disorders.
Dr Schweikert will additionally benefit from her continued association with the Cyber Valley, one of prime Europe’s centres for machine learning with a particular strength in causal inference.