Chemical reaction network and systems biology interface for scientific machine learning (SciML). High performance, GPU-parallelized, and O(1) solvers in open source software.
Is your feature request related to a problem? Please describe.
In the algebraic approach to mass action chemical reaction networks there is a very useful factorization of the RHS of the ODE system
$$
\frac{dx}{dt}=Y\cdot A_k \cdot \Phi(x),
$$
where $Y$ is the complex stoichiometric matrix, $A_k=-L_k$ and $\Phi(x)$ is the vector of monomials. In order to define $L_k$ consider the complex graph associated to the CRN as a weighted directed graph with reaction rates as weights and then $L_k$ is its Laplacian matrix.
While there is a function returning $Y$ I don't think there are functions for the other two factors.
Describe the solution you’d like
It will be great to have exposed functions for the last factors: $A_k$ and $\Phi(x)$
Describe alternatives you’ve considered
For $A_k$ it will be maybe enough to have the weighted graph of complexes.
As suggested here I am opening an issue.
Is your feature request related to a problem? Please describe.
In the algebraic approach to mass action chemical reaction networks there is a very useful factorization of the RHS of the ODE system $$ \frac{dx}{dt}=Y\cdot A_k \cdot \Phi(x), $$ where $Y$ is the complex stoichiometric matrix, $A_k=-L_k$ and $\Phi(x)$ is the vector of monomials. In order to define $L_k$ consider the complex graph associated to the CRN as a weighted directed graph with reaction rates as weights and then $L_k$ is its Laplacian matrix.
While there is a function returning $Y$ I don't think there are functions for the other two factors. Describe the solution you’d like
It will be great to have exposed functions for the last factors: $A_k$ and $\Phi(x)$
Describe alternatives you’ve considered
For $A_k$ it will be maybe enough to have the weighted graph of complexes.