Berlin 2012 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

BP: Fachverband Biologische Physik

BP 7: Posters: Proteins

BP 7.25: Poster

Monday, March 26, 2012, 17:30–19:30, Poster A

Using Dynamic Graphs to Quantitatively Visualize Agglomeration in Spatial Simulations — •Tihamer Geyer¹, Florian Lauck², and Volkhard Helms¹ — ¹Zentrum für Bioinformatik, Universität des Saarlandes, Saarbrücken — ²Dept. of Bioengineering and Therapeutic Sciences, UCSF, San Francisco, USA

The usual approaches to analyze many-particle simulations of association processes either focus on time-averaged measures for the degree of binding like radial distribution functions or cluster sizes, or movies are generated which in principle provide all the details but in an only qualitative way. Here we show how dynamic graphs can be used to visualize quantitatively the time dependent events of complex formation and breaking. For this, a simple distance criterion is used to set up a time dependent graph from the snapshots of the spatial simulation [1]. Some examples highlight how even simple graph measures like the degree distribution or the clustering coefficient can be used to follow the dynamics, to unambiguously identify complexes in a sea of monomers and partially assembled fragments, or to quantify how regular or amorphous an aggregate is [2]. In a further step the spatial simulation and the dynamic graph will be combined such that the simulation can make use of the connectivity encoded in the graph by, e.g., defining temporary pseudo-particles or different diffusion coefficients based on how many neighbors are actually bound.

[1] F. Lauck, V. Helms, T. Geyer, J. Chem. Theor. Comput. 5 (2009) 641

[2] T. Geyer, BMC Biophys. 4 (2011) 7