Jump to content


Photo
- - - - -

Turnover model with zero baseline and response is by subtraction

Turnover model

  • Please log in to reply
3 replies to this topic

#1 Pradeepsharma

Pradeepsharma

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 07 April 2019 - 12:35 AM

Anyone tried modelling turnover model using the  response data which is difference of drug treated and vehicle treated animals. Baseline in this case is zero. There is case study PD28 in Johan Gabrielsson/Dan Weiner book 'PK&PD data analysis: concepts and applications' which is similar to this. That case study describes Winnonlin code in it. Can anyone think of PML code equivalent of that code please? It appears that we need to have initial condition that need to be set by sequence statement and following is my try but now working:

 

 

deriv(E = Kin * (1 - Imax * C / (C + IC50)) - Kout * E-E0)
 
sequence{E=E0}
E0 = Kin / Kout
Kin=1

Edited by Pradeepsharma, 07 April 2019 - 12:49 AM.


#2 dlweiner

dlweiner

    Member

  • Members
  • PipPip
  • 23 posts

Posted 07 April 2019 - 12:29 PM

 

Anyone tried modelling turnover model using the  response data which is difference of drug treated and vehicle treated animals. Baseline in this case is zero. There is case study PD28 in Johan Gabrielsson/Dan Weiner book 'PK&PD data analysis: concepts and applications' which is similar to this. That case study describes Winnonlin code in it. Can anyone think of PML code equivalent of that code please? It appears that we need to have initial condition that need to be set by sequence statement and following is my try but now working:

 

 

deriv(E = Kin * (1 - Imax * C / (C + IC50)) - Kout * E-E0)
 
sequence{E=E0}
E0 = Kin / Kout
Kin=1

--------------

Here you go - see the change to the Observe statement

 

deriv(Aa = - K * Aa)
deriv(A1 = K * Aa - K * A1)
deriv(E = Kin * (1 - Imax * C ^ n / (C ^ n + ID50 ^ n)) - Kout * E)
dosepoint(Aa)
C = A1 / V
sequence{E = Kin / Kout}
error(EEps = 1)
observe(EObs = (E - Kin/Kout) + EEps)

  • Pradeepsharma likes this

#3 Pradeepsharma

Pradeepsharma

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 07 April 2019 - 06:18 PM

Many thanks Dr Weiner, I am able to fit the data now and it has worked well. 

 

It appears that once we change the observe statement, then for some reason, 'E vs t' or any of 'E' visualisation disappears from the 'initial estimates' window. 

 

This was not issue because I determined initial estimates based on theory described in your book and this is also explained in previous PML webinars of turnover model. 

 

Thanks again, much apprecaited.



#4 Pradeepsharma

Pradeepsharma

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 08 April 2019 - 10:23 AM

My apologies for coming back. I was thinking lately the need to fix Kin=1. Is it needed to fix Kin to fix to constant value (1) or is it parameter estimated along with Kout, Imax and IC50. Also I am keeping bounds for Imax as,  minimum value 0 to maximum value=1.

 

Any thoughts in this direction will be highly appreciated. 






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users