As you may have noticed, I rarely update this homepage to reflect my latest research topics. I have now started a blog. In contrast to most scientific blogs, I plan to primarily write about unpublished analyses and observations from my own research rather than about results that can already be found in the scientific literature.
Within my research area of bioinformatics, I have been working with a very wide variety of methods. There is, however, a common theme of my research, namely prediction of protein function. I believe that no single approach will be able to solve this problem in a foreseeable future; rather integration of many different types of evidence is needed for making predictions.
The main result of my Ph.D. work was a method for predicting various functional classes of eukaryotic proteins, ProtFun, which uses artificial neural networks to combine a large number of different sequence derived features. This work was done in Prof. Søren Brunak's group (CBS). A similar method, ArchaeaFun, adapted for predicting enzyme classes for archaeal proteins was also developed.
I am currently working as a staff scientist in Dr. Peer Bork's group at the European Molecular Biology Laboratory (EMBL). I mainly work on cross-species integration of large-scale experimental datasets in relation to the STRING, STITCH, and NetworKIN web resources. I was also involved in developing a method, DisEMBL, which allows structurally disordered segments of proteins to be predicted from sequence only.
Lately, my resarch has focused on understanding how cellular processes are regulated. This work has resulted in a a novel approach for unraveling signal-transduction pathways, NetworKIN, and in a new model for how transcriptional and posttranslational regulation jointly controls the dynamic assembly of protein complexes during the cell cycle.
For more information on my scientific work please refer to my Curriculum Vitae and List of Publications and Presentations.