Berlin 2008 – scientific programme
Parts | Days | Selection | Search | Downloads | Help
BP: Fachverband Biologische Physik
BP 25: Protein Structure and Folding
BP 25.7: Talk
Thursday, February 28, 2008, 16:00–16:15, PC 203
Protein structure prediction using structure profiles — •Katrin Wolff1, Andrea Cavalli2, Harri Hopeauroho2, Michele Vendruscolo2, and Markus Porto1 — 1Institut für Festkörperphysik, Technische Universität Darmstadt, Germany — 2Department of Chemistry, University of Cambridge, UK
Protein structure predictors using knowledge-based energy functions and homology modelling have faced huge progress over the last few years whilst still allowing for further improvement. Here, we investigate the possibility to improve an existing predictor by incorporating information from structure profiles. A specific class of such profiles can be used to reconstruct the protein structure’s contact matrix [1], from which the three-dimensional structure can be efficiently recovered [2]. Such profiles can be predicted to good accuracy from sequence (see e.g. [3]). It therefore seems promising to include profile information into existing predictors. Using the program collection ‘almost’ [4] we perform a series of prediction steps consisting of Monte Carlo-minimization with fragment insertion, filtering by the structure profile’s cost function and subsequent refinement of predicted structures. We compare conventionally predicted structures to those including the structure profile as an additional input for several known protein structures and show that RMSDs to the target structure significantly decrease.
[1] M. Porto et al., Phys. Rev. Lett. 92, 218101 (2004).
[2] M. Vendruscolo et al., Fold. & Des. 2, 295 (1997).
[3] A. R. Kinjo et al., BMC Bioinformatics 7, 401 (2006).
[4] A. Cavalli et al., Proc. Nat. Acad. Sci. 104, 9615 (2007)