Burkhard Rost

from Wikipedia, the free encyclopedia
Burkhard Rost 2013

Burkhard Rost (born July 11, 1961 ) is a German bioinformatician .

Rost went to school in Herzberg am Harz , did military service in the Air Force and from 1982 studied physics (and other subjects such as history, psychology and philosophy) at the University of Gießen and the University of Heidelberg , where he received his doctorate in 1994. He wrote his dissertation (Neural networks and evolution - advanced prediction of protein secondary structure) at EMBL , where he was also a post-doctoral student. He was also at the European Bioinformatics Institute (EBI) in Hinxton (Cambridgeshire) and for a short time at LION bioscience in Heidelberg. In 1998 he became an Assistant Professor and 2000 Associate Professor at Columbia University's Faculty of Medicine . In 2009 he received a Humboldt Professorship and became a professor at the Technical University of Munich .

He uses methods of machine learning ( neural networks , support vector machine ) for predictions, for example, of the secondary structure of proteins and their function, prediction of the effect of point mutations. He combines the methods with evolutionary information. He makes the software developed (such as the PredictProtein Internet server since 1992) partially accessible online. His group also pursues applications in medicine.

At the beginning of his career he worked as a theoretical physicist with spin glasses and neural networks as models for the brain. His diploma thesis in physics in Heidelberg was about learning algorithms for spin-glass-like neural networks . From 1988 to 1990, with funds from the Volkswagen Foundation , he investigated a theoretical project in the USA to verify arms control with seismic and acoustic sensor networks.

From 2007 to 2014 he was President of the International Society for Computational Biology (ISCB).

Fonts (selection)

  • with C. Sander: Prediction of protein secondary structure at better than 70% accuracy, J. Mol. Biol., Volume 232, 1992, pp. 584-599.
  • with C. Sander: Combining evolutionary information and neural networks to predict protein secondary structure, in: Proteins: Structure, Function, and Genetics, Volume 19, 1994, pp. 55-72.
  • PHD: predicting one-dimensional protein structure by profile-based neural networks, Methods in Enzymology, Volume 266, 1996, pp. 525-539.
  • Twilight zone of protein sequence alignments, Protein Engineering, Vol. 12, 1999, pp. 85-94
  • with G. Yachdav, J. Liu: The PredictProtein server, Nucleic acids research, Volume 32 (Suppl. 2), 2004, W321 – W326
  • with P. Caminci a. a .: The transcriptional landscape of the mammalian genome, Science, Volume 309, 2005, pp. 1559-1563
  • with A. Schlesinger, J. Liu: Natively unstructured loops differ from other loops, PLoS Computational Biology. 2007; 3: e140.
  • with Y. Bromberg: SNAP: predict effect of non-synonymous polymorphisms on function, Nucleic Acids Res., Volume 35, 2007; Pp. 3823-3835.
  • with R. Nair: Mimicking cellular sorting improves prediction of subcellular localization, J. Mol. Biol., Volume 348, 2005, pp. 85-100.
  • with Y. Ofran: Protein-protein interaction hotspots carved into sequences, PLoS Computational Biology. 2007; 3: e119.

Web links