Rune Lyngsø

Professional

I received an MSc in computer science and mathematics from the University of Aarhus in 1997, and a PhD based on my dissertation entitled Computational Biology from the same institution in 2000. Since then I have been working first in the group of David Haussler and now in the group of Jotun Hein.

Though I have spent more than a decade in the field of bioinformatics, I still very much classify myself as a computer scientist. The focus of my work has been development of algorithms and software for and understanding the complexity of problems from molecular biology. So from a computational point of view, my work can easily and succinctly be categorised as algorithms and complexity theory.

As a consequence, I have never had a favourite biological problem driving my research – rather it has been a case of discovering, or even stumbling on, bioinformatics problems that were computationally challenging and where I was capable of contributing. This means that I have been working in areas of bioinformatics as diverse as RNA secondary structure, alignment, single sequence analysis (probably more recognisable as the area of string algorithms), recombination inference in population data, microarray normalisation, and stochastic models of formal languages (i.e. hidden Markov models and stochastic context free grammars).

Despite having the computer science aspects close at heart, I soon realised that in an applied field like bioinformatics, there are people who need to solve concrete instances of a problem and are not just interested in the abstract problem (this after twice being confronted with Nice idea, has it been implemented yet? when presenting an algorithm). And who better to implement an algorithm than the computer scientist that came up with it? So over the years I've developed software for RNA secondary structure prediction, comparing an HMM with an SCFG, fast and robust normalisation of microarray data, inference and analysis of recombination in SNP data, and been involved with software for comparing two HMMs and Markov chain Monte Carlo sampling of statistical alignments. Recently I have even forayed in to analysis of the evolution of RNA genes with conserved structure.

Personal

When not working as a postdoc in Jotun Hein's group, I spend quite a lot of time playing volleyball in and around Oxford. From time to time my wife and I also benefit from the wonderful network of public rights of way in the UK, something I've not seen parallelled anywhere else.

Contact