Driven by data
As a physicist Søren Brunak was trained in the basic understanding of the tiniest parts of the universe – and not least their importance for the greater picture. In his current research, he surfs all scales by combining molecular data about our body’s building blocks with clinical data to form gigantic meta-studies that can predict the progression of illness in an entire population.
57-year-old Søren Brunak is a professor at The Novo Nordisk Foundation Centre for Protein Research at the University of Copenhagen. He is a Master of Science in Physics, but was one of the early birds to engage in trendy interdisciplinary activities as his Master’s thesis was partly realised at the Department of Neurophysiology at the University of Copenhagen. In 1991, Søren Brunak obtained his PhD in computational biology at the Technical University of Denmark, where he is still affiliated in the capacity of professor.
As a pioneer in bioinformatics, Søren Brunak was one of the first scientists in Denmark to introduce the so-called neutral network – computer models inspired by our brain structure, which turn into algorithms based on a huge amount of data input.
Why is your research important to society?
There’s been rapid progress in data driven research within the last few years. First and foremost because accessing data via industrialised experimental methods has become much cheaper. The mapping of the first human genome cost DKK 20 billion, while today, DNA sequencing has evolved into streamlined perfection. Biological research of earlier times was most often driven by a few dedicated people in white coats in a lab who broke test tubes and wrote reports – today, we’re talking huge international teams who have computers with tremendous calculative powers to assist them in data analyses.
My research is carried out on some sort of meta-level, where it’s all about combining molecular data with clinical data in order to gain new knowledge about the development of illnesses and diseases in the individual human being. Our analyses take their point of departure in Integrative Systemic Biology, which enables us to spot the general disease patterns in a badly arranged data-chaos.
In the long term, we’re heading in the direction of individualised treatment, i.e. treatment that takes the individual patient’s genetic structure into account. Being a doctor would be so much easier if we were all clones, however, all our genes vary and in principle, all illnesses and diseases are variations on a specific genetic background. The perspective is that bioinformatics will feed us the information need to reveal the development of illnesses and health throughout a lifetime. As a society, we can save a lot of money if we’re able to provide people with a more targeted and thus efficient treatment.
Tell us about a career high for you
Our research area was surrounded by a great amount of scepticism in the early days – both from biologists and not least the computer science research environments. People did not want to believe that a computer with a data-driven algorithm in and of itself could reveal anything worthwhile about new data. The preference was to base such findings on logic deductions: If we know this and that about a specific protein, we can deduct that it will do such and such.
We were among the first to predict the zip code of proteins – the so-called signal peptides – by way of neutral networks and so-called machine learning, however, the method and our results were not immediately and generally accepted. These days though, our scientific articles on the subject have been quoted in excess of 10,000 times.
When and why did you choose to become a scientist?
I always liked going to school, and therefore it was quite clear to me from an early stage that I wanted an academic education. My parents were, respectively, schoolteacher and financial executive, so a career as a researcher was perhaps not the most obvious choice, even if their circle of friends included several scientists. Versatility is healthy – my father’s father was a dairyman, my eldest son is a cabinet-maker. The professions in our family span far and wide. Personally, I painted greenhouses for professional gardeners while studying – and I was actually quite good at it.
Why did you choose a university career?
As a researcher at a university you have great freedom and can occupy yourself with what you find most interesting. However, I’ve also been in close contact with the industry during my career – I have, for example, sat on many a scientific advisory board in specific businesses. The methods we’ve helped develop have more than 30 million annual page views on the Internet. Researchers in their tens of thousands use them. You could say that it’s a kind of e-trade, only without profit. On the other hand, the university gains tremendously in terms of citations and professional influence.
How do you spend your leisure time?
I have four children aged 13, 16, 20 and 30. I’m married to Pernelle and we live in Ryvangskvarteret, quite close to my childhood home by Emdrup Lake. My wife is a painter and textile designer. Aesthetics and interior decorating play – not least on account of my wife – a significant part in my private life. I love scanning flea markets both at home and abroad. For a while I collected old American coffee machines, but usability is not a must in my flea market finds. I also have a soft spot for Mercedes cars of a (I guess) vintage standing. When I have time, I like to repair and restore both our house and summerhouse – manual work helps me relax.