Skip to main content

Table 3 Experiments for reducing prediction uncertainty

From: Optimal experiment selection for parameter estimation in biological differential equation models

Iteration

Perturbation

Measurement

Estimated error

1

Wild

Microarray

7.7×1010

2

Down 5

Proteins 2 and 6

3.5×102

3

Delete 5

Proteins 4 and 6

2.5

4

Over 4

Proteins 4 and 5

2.4×10−1

  1. We list the sequence of experiments for estimating the predictions for several protein concentrations in model 1. At each iteration we reduce the relative uncertainty in the predictions given by Eq. (13) which is reported in the estimatd error column. For this measure of error, a value of 1 corresponds to an accuracy comparable to the experimental noise. Notice that the uncertainty in the predictions can be reduced to the experimental uncertainty with just four experiments. Even though the parameters are largely unconstrained after four experiments, the model is nevertheless able to make falsifiable predictions.