Steadystate solver fails if preequilibration starts in steadystate

[dweindl]

E.g. preequilibration for the conversion reaction example with initial states [0, 0] fails:

    solver.setSensitivityMethod(amici.SensitivityMethod_forward)
    solver.setSensitivityOrder(amici.SensitivityOrder_first)
    model.setSteadyStateSensitivityMode(
        SteadyStateSensitivityMode_simulationFSA)

Steady state sensitivity computation failed due to unsuccessful factorization of RHS Jacobian

preeq_status: [[1 0 0]]

preeq_numsteps [[0 0 0]]

x_ss and sx_ss seem to be set correctly, though.

[FFroehlich]

E.g. preequilibration for the conversion reaction example with initial states [0, 0] fails:

    solver.setSensitivityMethod(amici.SensitivityMethod_forward)
    solver.setSensitivityOrder(amici.SensitivityOrder_first)
    model.setSteadyStateSensitivityMode(
        SteadyStateSensitivityMode_simulationFSA)

Steady state sensitivity computation failed due to unsuccessful factorization of RHS Jacobian

preeq_status: [[1 0 0]]

preeq_numsteps [[0 0 0]]

x_ss and sx_ss seem to be set correctly, though.

Is it possible that the Jacobian is singular in that case, which (correctly) prevents computation of sensitivities?

[paulstapor]

It is indeed singular, but the fun thing is: Simulation started in steady state. So all the steadystate solver would have to do is just output sx0...

[FFroehlich]

It is indeed singular, but the fun thing is: Simulation started in steady state. So all the steadystate solver would have to do is just output sx0...

Good point!

[paulstapor]

Will do on Friday, hopefully. Should be easy to fix

[elbaraim]

I have observed this behavior also in the Raimundez_PCB2020 model.

( already discussed with @paulstapor , I just wanted to add here for documentation )