Indirect_PKPD_model

[Generated automatically as a Tutorial summary]

Model Description

Name:

indirect_pd_pop

Title:

Indirect_PKPD_model

Author:

PoPy for PK/PD

Abstract:

A indirect (i.e. uses delay compartments) PD Model, based on the amount of drug in the body, delayed by two lag compartments
The amount in the central compartment is determined by K, which has been estimated for each individual.
The amount in the central compartment influences the rate of removal of a biomarker (KOUT).
Keywords:

pd; one compartment model; indirect; delay compartment

Input Script:

indirect_pd_tut.pyml

Diagram:

Comparison

True objective value

3940.3541

Final fitted objective value

3933.3780

Compare Main f[X]

Name

Initial

Fitted

True

Abs. Error

Prop. Error

f[BASE]

500

782

800

1.81e+01

2.26%

f[KOUT]

0.1

0.0286

0.03

1.36e-03

4.52%

Compare Noise f[X]

Name

Initial

Fitted

True

Abs. Error

Prop. Error

f[PNOISE]

0.05

0.0967

0.1

3.33e-03

3.33%

f[ANOISE]

0.2

0.201

0.5

2.99e-01

59.81%

Compare Variance f[X]

Name

Initial

Fitted

True

Abs. Error

Prop. Error

f[BASE_isv]

0.02

0.0345

0.05

1.55e-02

30.95%

f[BASE_isv;KOUT_isv]

0

0.0047

0

4.70e-03

inf

f[KOUT_isv;BASE_isv]

0

0.0047

0

4.70e-03

inf

f[KOUT_isv]

0.02

0.0407

0.01

3.07e-02

306.95%

Outputs

Fitted f[X] values (after fitting)

f[BASE] = 781.9491
f[KOUT] = 0.0286
f[PNOISE] = 0.0967
f[ANOISE] = 0.2009
f[BASE_isv,KOUT_isv] = [
    [ 0.0345, 0.0047 ],
    [ 0.0047, 0.0407 ],
]

Generated data .csv file

Synthetic Data:

synthetic_data.csv

Gen and Fit Summaries

Inputs

True f[X] values (for simulation)

f[BASE] = 800.0000
f[KOUT] = 0.0300
f[PNOISE] = 0.1000
f[ANOISE] = 0.5000
f[BASE_isv,KOUT_isv] = [
    [ 0.0500, 0.0000 ],
    [ 0.0000, 0.0100 ],
]

Starting f[X] values (before fitting)

f[BASE] = 500.0000
f[KOUT] = 0.1000
f[PNOISE] = 0.0500
f[ANOISE] = 0.2000
f[BASE_isv,KOUT_isv] = [
    [ 0.0200, 0.0000 ],
    [ 0.0000, 0.0200 ],
]