Fig. 1
From: Classification of protein–protein association rates based on biophysical informatics
The flowchart of the overall computational procedure. A coarse-grained Monte-Carlo simulation was used to generate a large number of protein–protein association trajectories (a). Based on counting how many encounter complexes were formed among all the trajectories, the association probability can be derived. Simulations are further carried out under different values of distance cutoff, so that a profile of association probabilities was generated for each protein complex in the benchmark set. Using these association profiles as input, a back propagation neural network algorithm was then applied to classify whether the association rate of a protein complex is higher than a predefined threshold (b)