StrikePy: Nonlinear observability analysis of inputs, states, and parameters in Python.

Abstract

[Abstract] Dynamic models typically have unknown parameters that must be estimated from data, and they may also have unknown inputs (disturbances). The concept of observability, which describes the possibility of inferring the internal state of a system by measuring its output, can be extended to account also for the possibility of inferring its unknown parameters and inputs. Such an extension leads to a property that may be called FISPO (Full Input, State, and Parameter Observability). Its analysis is particularly relevant in systems biology, since models from this area often have a large number of unknown parameters, as well as state variables that cannot be measured due to experimental limitations. It is usually challenging to assess the FISPO of nonlinear models, which has motivated the development of specialized software such as the MATLAB toolbox STRIKE-GOLDD. However, despite the increasing popularity of Python among the biological modelling community, there was a lack of computational tools for FISPO analysis in this language. To fill this gap, we have developed an open source software toolbox, StrikePy, which implements the core functionalities in STRIKE-GOLDD.

Ministerio de Ciencia e Innovación; 10.13039/501100011033