technical report Crosstalk in oscillatory pathways: Frequency Domain Analysis


Constructing meaningful and useful interacting networks of reactions from disparate experimental results is one of the greatest challenges facing systems biology. We present a methodology which attempts to address this challenge, using coupled oscillatory models of p53, NF-kB and the mammalian cell cycle as our focus. Starting with recent state-of-the-art models of these systems, informed by online databases of protein networks and interactions, their key elements are distilled into simplified models containing the most significant parts. Parameters for the individual and coupled models are estimated in a deterministic modelling framework and these parameters are then used to construct the final stochastic models used in our analysis. We present a new frequency domain-based analysis technique which is particularly relevant for such oscillatory systems and is most revealing when applied to the results of stochastic simulation.

Using our techniques, we show that the interactions between the networks we have considered are highly complex and not intuitive; points of maximum perturbation do not necessarily correspond to points of maximum proximity to influence and the quality of perturbation is frequency dependent.

Paper Details


S. Sedwards,  A. Ihekwaba