Add support for parameter estimation

[tomvej]

Parasim so far supports mainly analysis of initial condition (i.e. simulation from different concentrations). However, in systems biology, parameter estimation and corresponding robustness analysis is more common.

Apparently, setting parameters as species does not work. Try running SIR model.
Octave seems unable to simulate species not present in reactions.

I have tried including "parameter species" in meaningless reaction (no change). Octave then has problems with convergence.

[tomvej]

Copasi way: SBML species has argument constant which marks it is not simulated by reactions.

This may be a way of introducing parameter estimation without being bogged down by everything in listOfParameters. However, current ODE model does not recognize the constant argument.