High Fidelity Modeling of Pulse Dynamics using Logic Networks

Pulsatile dynamics are a key regulatory behavior in many biological systems, but perform poorly using conventional logic network methods. This work addresses the relatively poor performance of logic network models when applied to pulse dynamic scenarios by implementing a novel update scheme. Models from this novel update scheme and a traditional update scheme are compared using canonical examples, i.e. the van der Pol oscillator and the Hypothalamic-Pituitary-Gonadal (HPG) axis. By incorporating this new update scheme, the authors show a 35% (11%) reduction of global minimum error in the van der Pol model (HPG axis) when compared with traditional methods.