Prediction of debacle parts for robustness in a cell by using recurrent neural networks

Living organisms have sophisticated control mecha,nism to keep biological system robust against abnormalities from inside/outside of them. However, at the same time, the control mechanism has a critical point at which the stability can be broken easily. This paper proposes a method to find critical points of the control mechanism in a biological pathway described by hybrid functional Petri nets (HFPN). In this method. HFPNs are converted to a recurrent neural networks (RNNs), checking robustness of the biological pathway with the RNN, a^nd finding some crucial points for the robustness. An example to apply this method to an apoptosis pathway is also presented.

Petri Net

Genomic Object Net

Recurrent Neural Networks

BPTT