Identifiability of Stochastically Modelled Reaction Networks

TL;DR

This paper investigates how certain types of data from stochastic biochemical systems can uniquely determine the underlying reaction network structure and parameters, enhancing understanding of system identifiability.

Contribution

It introduces conditions under which the network structure and parameters are identifiable from stochastic dynamics data, and evaluates inference accuracy through simulations.

Findings

Certain data types can uniquely identify network structure and parameters.

Network inference accuracy improves with more detailed stochastic data.

Theoretical conditions for identifiability are established.

Abstract

Chemical reaction networks describe interactions between biochemical species. Once an underlying reaction network is given for a biochemical system, the system dynamics can be modelled with various mathematical frameworks such as continuous time Markov processes. In this manuscript, the identifiability of the underlying network structure with a given stochastic system dynamics is studied. It is shown that some data types related to the associated stochastic dynamics can uniquely identify the underlying network structure as well as the system parameters. The accuracy of the presented network inference is investigated when given dynamical data is obtained via stochastic simulations.