log_likelihood: User defined likelihood function.
In pauwels2014: Bayesian Experimental Design for Systems Biology.

Description Usage Arguments Details Value Author(s) Examples

Noise is assumed to be independent for each entry. The default likelihood assumes the same heteroscedastic noise as the model used in add_noise.

1	log_likelihood(simu, simu_subset, data, fit = F)

`simu`	Simulated time courses.
`simu_subset`	Subset of the simulated time course which relates to observed data.
`data`	Observed data.
`fit`	Should smoothness prior information about simulation added to the prior or not?

The noise model is gaussian with variance of the form (0.01 + 0.04 * m^2) where m is the mean.

A numerical value.

Edouard Pauwels

data(experiment_list1)
data(observables)

## Generate the knowledge object with correct parameter value
knobj <- generate_our_knowledge(transform_params)

simu <- simulate_experiment(knobj$global_parameters$true_params_T, knobj, experiment_list1$nothing)

simu_subset <- simu[
 knobj$global_parameters$tspan %in% observables[["mrnaLow"]]$reso, 
 observables[["mrnaLow"]]$obs
]

data <- add_noise(simu_subset)

log_likelihood(simu, simu_subset, data)