We are experiencing a minor revolution with the growth of big data, according to Professor of Mathematics Magnus Fontes at LTH. He therefore recently took the initiative to establish the Centre for Mathematical Modelling, to gather the University’s expertise in big data modelling and calculations. The idea is that less initiated researchers will also be able to receive support from colleagues with experience of issues concerning modelling and analysis of big data. It will also be easier for colleagues, funding bodies and researchers at other universities to find and contact those with appropriate expertise.
“Progress is very rapid. Just take the human genome, which was mapped around the millennium. It took ten years and cost USD 3 billion in labour, chemicals and hardware. Today it costs less than USD 1 000 to sequence a human genome and it only takes a day or two. In a few years I think it will take a few hours and cost 20 dollars”, says Magnus Fontes, who is currently a visiting professor at the Institut Pasteur in Paris.
The accelerating technological possibilities lead to diseases increasingly being treated on the basis of the patient’s genetic and phenotypic characteristics, and their medical treatment can be adapted accordingly.
“There is an incredible amount happening in the field of biomedicine at the moment. We can compare it with astronomy, in which the lion’s share of the work has long been done using calculations and computer simulations. This shift from a descriptive to a modelling science can also be expected in biology and biomedicine”, he predicts.
New technology requires new skills. Those who mostly look into a microscope and culture cells in vitro may not need any profound knowledge of maths. However, if you instead spend your days using advanced computer software to analyse and visualise patients’ genetic expression, proteins and their interaction with bacteria, viruses and parasites, you will need a little more insight into how calculations and statistics work – not at specialist level, but so that you can be a good customer, user and also research colleague if you are part of interdisciplinary projects.
“Many biomedical research institutes, for instance the Institut Pasteur, are investing more and more in training biomedicine students in both statistics and numerical analysis. This is to increase the quality of dialogue between mathematic modellers, statisticians and researchers in numerical analysis”, he explains.
Magnus Fontes thinks this knowledge should also be incorporated to a greater extent in the education offered at Lund University. An American report commissioned by President Obama (McKinsey 2011) predicted a huge shortage of big data analysts in the public and private sectors in just a few years.
“As a first step, doctoral students in biomedicine should have greater opportunities to improve their general knowledge of numerical analysis, statistics and basic modelling.”
However, equipping students at home in Lund doesn’t have to be the only solution. The Centre for Mathematical Sciences has always had close interaction with both local universities and leading institutions elsewhere. The department’s doctoral students are welcome to attend their courses and their doctoral students are welcome here.
“We have a number of excellent research environments in Lund that make us attractive for research and doctoral student exchanges. The foundation for successful research is advanced doctoral student programmes. This network model has worked well for us”, he says.
In Paris at the moment, Magnus Fontes is accompanied by three Lund doctoral students who are collaborating with researchers worldwide on various international projects.
The internationalisation of research is a strong trend in academia. Many projects are carried out in research networks spanning several continents, which take in the expertise they need and open doors for broad international collaborations.
“This is especially the case in modelling and analysis of big data in biomedicine”, says Magnus Fontes.
When data processing shifts from tiny to elephantine proportions, things that were previously reserved for the research lab become public property.
“A Stanford researcher has sequenced parts of his genome over four years. In that way, he can see if he is getting a cold before he develops any symptoms. He can then cancel next week’s meeting in good time. The technology for this is not really very advanced, but it is expensive. In a few years, we will see much more of this kind of thing”, says Magnus Fontes.
Text: Kristina Lindgärde
Photo: Kennet Ruona