Dear users,
I am trying to compareseveral specific PK/PD study designs by:
-- run 200 simulations with the final model (develope from original dataset)
-- fit the final model to the 200 simulated dataset
To achieve above, I used $SIM SUBPROB 0 option
however, nonmem would completely stop afterrunning intoestimationproblem atone specific simulation/estimation cycles, for some designs it stop even before 10th iterations.
Is there anyway nonmem could continue go on?
Or, does someone know alternative way to achieve the goal?
thanks
simulation question
You can use S+ (or R that has an advantage of being free) to control
nonmem, but you may need a bit of programming. One possible scheme is to
prepare template control stream (with $SIM SUBPROB=1 option) and then
run the loop
for(i in 1:NumerOfIterations) {
# prepare control file from a template
# (need to change random seed in the $SIM record)
system(NonmemCOmmandToRunNonmem)
}
The control stream should include INFN routine or INFN block that
outputs parameter estimates to some file (adds them to the file, not
over-writes it). R (or S+) working directory should contain all the
files needed to run nonmem (in R, use setwd(NonmemDirName) to do it)
Leonid
--------------------------------------
Leonid Gibiansky, Ph.D.
President, QuantPharm LLC
web: www.quantpharm.com
e-mail: LGibiansky at quantpharm.com
tel: (301) 767 5566
Ethan Wu wrote:
> Dear users,
> I am trying to compare several specific PK/PD study designs by:
> -- run 200 simulations with the final model (develope from original
> dataset)
> -- fit the final model to the 200 simulated dataset
> To achieve above, I used $SIM SUBPROB=200 option
> however, nonmem would completely stop after running
> into estimation problem at one specific simulation/estimation cycles,
> for some designs it stop even before 10th iterations.
> Is there anyway nonmem could continue go on?
> Or, does someone know alternative way to achieve the goal?
> thanks
>
>
Dear users,
I am trying to compare several specific PK/PD study designs by:
-- run 200 simulations with the final model (develope from original dataset)
-- fit the final model to the 200 simulated dataset
To achieve above, I used $SIM SUBPROB=200 option
however, nonmem would completely stop after running into estimation problem
at one specific simulation/estimation cycles, for some designs it stop even
before 10th iterations.
Is there anyway nonmem could continue go on?
Or, does someone know alternative way to achieve the goal?
thanks
You can use S+ (or R that has an advantage of being free) to control nonmem, but you may need a bit of programming. One possible scheme is to prepare template control stream (with $SIM SUBPROB=1 option) and then run the loop
for(i in 1:NumerOfIterations) {
# prepare control file from a template
# (need to change random seed in the $SIM record)
system(NonmemCOmmandToRunNonmem)
}
The control stream should include INFN routine or INFN block that outputs parameter estimates to some file (adds them to the file, not over-writes it). R (or S+) working directory should contain all the files needed to run nonmem (in R, use setwd(NonmemDirName) to do it)
Leonid
--------------------------------------
Leonid Gibiansky, Ph.D.
President, QuantPharm LLC
web: www.quantpharm.com
e-mail: LGibiansky at quantpharm.com
tel: (301) 767 5566
Ethan Wu wrote:
> Dear users, I am trying to compare several specific PK/PD study designs by: -- run 200 simulations with the final model (develope from original dataset)
>
> -- fit the final model to the 200 simulated dataset
> To achieve above, I used $SIM SUBPROB=200 option
>
> however, nonmem would completely stop after running into estimation problem at one specific simulation/estimation cycles, for some designs it stop even before 10th iterations.
>
> Is there anyway nonmem could continue go on?
> Or, does someone know alternative way to achieve the goal?
> thanks
Dear Ethan,
the simplest solution to solve your problem would be to use the bootstrap
command of PSN ( http://psn.sourceforge.net/). With PSN installed you would
simply do:
bootstrap final_model.mod -samples=200
PSN would take care of unsuccessful runs and provide a nice summary of the
individual estimates.
Best regards
Sebastian
Quoted reply history
On Thu, Feb 5, 2009 at 11:18 PM, Ethan Wu <[email protected]> wrote:
> Dear users,
> I am trying to compare several specific PK/PD study designs by:
> -- run 200 simulations with the final model (develope from original
> dataset)
> -- fit the final model to the 200 simulated dataset
> To achieve above, I used $SIM SUBPROB=200 option
> however, nonmem would completely stop after running
> into estimation problem at one specific simulation/estimation cycles, for
> some designs it stop even before 10th iterations.
> Is there anyway nonmem could continue go on?
> Or, does someone know alternative way to achieve the goal?
> thanks
>
>
>
Dear Ethan,
Perl-speaks-NONMEM (PsN) is a free program developed by Uppsala Research
Pharmacometrics group. We recently added a module performing Stochastic
Simulations followed by re-Estimations (SSE). The purpose is to investigate
aspects related to simulation or estimation methods, models, study designs
or statistical inference. I think it will solve your problem.
You can download it here as well as find the sse userguide:
http://psn.sourceforge.net/
What it does compared to NSUB=X is to run all problems in parallel, so a
termination won't affect the others. And moreover it provides a summary with
computations like bias etc... A simple single command will do the simulation
and the re-estimation with the different designs: sse run1.mod -samples=200
-alt=run2.mod
You can contact me or my colleagues for more details.
Best regards,
Elodie
Elodie L. Plan, PharmD, MSc,
PhDstudent
*******************************************
Div. of Pharmacokinetics and Drug Therapy,
Department of Pharmaceutical Biosciences,
Faculty of Pharmacy, Uppsala University
Box 591 - 751 24 Uppsala - SWEDEN
Office +46 18 4714385 - Fax +46 18 4714003
----------------------------------------------------
Quoted reply history
From: [email protected] [mailto:[email protected]] On
Behalf Of Ethan Wu
Sent: Thursday, February 05, 2009 11:19 PM
To: [email protected]
Subject: [NMusers] simulation question
Dear users,
I am trying to compare several specific PK/PD study designs by:
-- run 200 simulations with the final model (develope from original
dataset)
-- fit the final model to the 200 simulated dataset
To achieve above, I used $SIM SUBPROB=200 option
however, nonmem would completely stop after running into estimation
problem at one specific simulation/estimation cycles, for some designs it
stop even before 10th iterations.
Is there anyway nonmem could continue go on?
Or, does someone know alternative way to achieve the goal?
thanks
Ethan,
Sebastian is right that a non-parametric bootstrap may be suitable for
determining the uncertainty in the population parameters. However, I got
the impression that you wanted to investigate possible study designs on
how informative they are for a future model-based analysis? If you would
like to do simulation based on your current best guess of the model
parameters (i.e. the point estimates) and you would like to do model
based analysis of the future study in isolation, then PsN has another
program which is highly efficient and would automatically provide you
with summary statistics, without any programming required. This program
is called sse for stochastic simulation and estimation:
http://psn.sourceforge.net/PDF_docs/sse_userguide.pdf
sse would also allow you to investigate the performance of alternative
models, e.g. simulation with a two-compartment model and estimation with
a one-compartment model, to see if CL can be estimated with good
precision and low bias, even from the sparse data of a future study.
You need to generate the data sets with the study designs yourself, but
the rest is really slick. If you are planning to analyse the new study
in conjunction with the currently-available data (i.e. a pooled
analysis) there may be some clever way of tweaking sse into evaluating
this, but I could not say exactly how to best achieve that. (maybe
someone in Uppsala has a suggestion in that case)
Best regards
Jakob
Quoted reply history
________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of Sebastian Ueckert
Sent: 06 February 2009 09:51
To: Ethan Wu
Cc: [email protected]
Subject: Re: [NMusers] simulation question
Dear Ethan,
the simplest solution to solve your problem would be to use the
bootstrap command of PSN ( http://psn.sourceforge.net/). With PSN
installed you would simply do:
bootstrap final_model.mod -samples=200
PSN would take care of unsuccessful runs and provide a nice summary of
the individual estimates.
Best regards
Sebastian
On Thu, Feb 5, 2009 at 11:18 PM, Ethan Wu <[email protected]> wrote:
Dear users,
I am trying to compare several specific PK/PD study designs by:
-- run 200 simulations with the final model (develope from original
dataset)
-- fit the final model to the 200 simulated dataset
To achieve above, I used $SIM SUBPROB=200 option
however, nonmem would completely stop after running into estimation
problem at one specific simulation/estimation cycles, for some designs
it stop even before 10th iterations.
Is there anyway nonmem could continue go on?
Or, does someone know alternative way to achieve the goal?
thanks
Dear all,
PsN is useful tool for this purpose but I think that the functionality that
should be used is called "Stochastic Simulation and Estimation (SSE)". The
Bootstrap functionality is primarily used for obtaining parameter
uncertainties. A detailed userguide for the SSE functionality is presented
at the PsN webpage (see link below).
http://psn.sourceforge.net/PDF_docs/sse_userguide.pdf
Write to the PsN mailing list if you experience any troubles and need help
([email protected]).
Best regards,
Martin
Quoted reply history
From: [email protected] [mailto:[email protected]] On
Behalf Of Sebastian Ueckert
Sent: Friday, February 06, 2009 10:51
To: Ethan Wu
Cc: [email protected]
Subject: Re: [NMusers] simulation question
Dear Ethan,
the simplest solution to solve your problem would be to use the bootstrap
command of PSN ( http://psn.sourceforge.net/). With PSN installed you would
simply do:
bootstrap final_model.mod -samples=200
PSN would take care of unsuccessful runs and provide a nice summary of the
individual estimates.
Best regards
Sebastian
On Thu, Feb 5, 2009 at 11:18 PM, Ethan Wu <[email protected]> wrote:
Dear users,
I am trying to compare several specific PK/PD study designs by:
-- run 200 simulations with the final model (develope from original
dataset)
-- fit the final model to the 200 simulated dataset
To achieve above, I used $SIM SUBPROB=200 option
however, nonmem would completely stop after running into estimation
problem at one specific simulation/estimation cycles, for some designs it
stop even before 10th iterations.
Is there anyway nonmem could continue go on?
Or, does someone know alternative way to achieve the goal?
thanks
Ethan,
As Leonid suggests, you may want to control NONMEM runs using an external tool and an INFN routine for extraction of results. MIfuns, an open-source R package, provides functions for this sort of automation. Template INFN codes are provided as well. It's a free download:
http://metruminstitute.org/downloads/index.shtml
R is good choice for this purpose because you can also script tasks such as data preparation, post processing of output and graphics and table creation - all within the same language. Other options for this sort of automation use a variety of scripting languages. Some examples are: Perl-Speaks NONMEM (PSN), Wings for NONMEM (WFN) and others... see the PAGE website for links to these projects ( http://www.page-meeting.org/ ).
Good luck,
Marc
Marc R. Gastonguay, Ph.D.
President & CEO, Metrum Research Group LLC [www.metrumrg.com]
Scientific Director, Metrum Institute [www.metruminstitute.org]
Direct: 860-670-0744 Main: 860-735-7043
Email: [email protected]
Quoted reply history
On Feb 5, 2009, at 5:18 PM, Ethan Wu wrote:
> Dear users,
> I am trying to compare several specific PK/PD study designs by:
>
> -- run 200 simulations with the final model (develope from original dataset)
>
> -- fit the final model to the 200 simulated dataset
> To achieve above, I used $SIM SUBPROB=200 option
>
> however, nonmem would completely stop after running into estimation problem at one specific simulation/estimation cycles, for some designs it stop even before 10th iterations.
>
> Is there anyway nonmem could continue go on?
> Or, does someone know alternative way to achieve the goal?
> thanks
Hi,
I dont understand why this wasn't fixed in NONMEM VI. It was a well known bug in NONMEM V and has caused lots of grief for users.
I hope the NONMEM VII developers have got this high on their list of NONMEM bugs that need to be fixed.
Nick
> > Dear users, I am trying to compare several specific PK/PD study designs by: -- run 200 simulations with the final model (develope from original dataset)
> >
> > -- fit the final model to the 200 simulated dataset
> > To achieve above, I used $SIM SUBPROB=200 option
> >
> > however, nonmem would completely stop after running into estimation problem at one specific simulation/estimation cycles, for some designs it stop even before 10th iterations.
> >
> > Is there anyway nonmem could continue go on?
> > Or, does someone know alternative way to achieve the goal?
> > thanks
--
Nick Holford, Dept Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New Zealand
[email protected] tel:+64(9)923-6730 fax:+64(9)373-7090
http://www.fmhs.auckland.ac.nz/sms/pharmacology/holford
Yes, it is fixed in Nonmem VII.
Katya
--------------------------
Ekaterina Gibiansky
Senior Director, PKPD, Modeling & Simulation
ICON Development Solutions
[email protected]
Quoted reply history
-----Original Message-----
From: [email protected] [mailto:[email protected]]
On Behalf Of Nick Holford
Sent: Friday, February 06, 2009 6:39 AM
To: [email protected]
Subject: Re: [NMusers] simulation question
Hi,
I dont understand why this wasn't fixed in NONMEM VI. It was a well
known bug in NONMEM V and has caused lots of grief for users.
I hope the NONMEM VII developers have got this high on their list of
NONMEM bugs that need to be fixed.
Nick
>> Dear users,
>> I am trying to compare several specific PK/PD study designs by:
>> -- run 200 simulations with the final model (develope from original
>> dataset)
>> -- fit the final model to the 200 simulated dataset
>> To achieve above, I used $SIM SUBPROB=200 option
>> however, nonmem would completely stop after running
>> into estimation problem at one specific simulation/estimation cycles,
>> for some designs it stop even before 10th iterations.
>> Is there anyway nonmem could continue go on?
>> Or, does someone know alternative way to achieve the goal?
>> thanks
>>
>>
>>
>
--
Nick Holford, Dept Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New
Zealand
[email protected] tel:+64(9)923-6730 fax:+64(9)373-7090
http://www.fmhs.auckland.ac.nz/sms/pharmacology/holford
Hi Ethan,
If think you get the message "message issued from estimation step" you still
have the $ESTIMATION block in your control stream, although you're using
SIMONLY option. Try removing the $ESTIMATION block and try again. Perhaps you
could post your control stream as well?
Sincerely,
Rob
_______________________________
Rob ter Heine, MSc, PharmD
Department of Pharmacology, Slotervaart Hospital
Amsterdam, The Netherlands
E: [email protected]
T: +31-20-5124737
>>> Ethan Wu <[email protected]> 02/06/09 2:28 >>>
Hi Robm
the simulatio step runs fine as I checked by using SIMONLY
the termination says
"0PROGRAM TERMINATED BY OBJ
ERROR IN NCONTR WITH INDIVIDUAL 16 ID=0.16000000E+02
NUMERICAL HESSIAN OF OBJ. FUNC. FOR COMPUTING CONDITIONAL ESTIMATE
IS NON POSITIVE DEFINITE
MESSAGE ISSUED FROM ESTIMATION STEP
AT 0TH ITERATION, UPON EVALUATION OF GRADIENT WITH RESPECT TO OMEGA
0TRY SUBSTANTIALLY INCREASING "INITIAL ESTIMATE OF SIGMA"
OR DECREASING "INITIAL ESTIMATE OF OMEGA""
Quoted reply history
________________________________
From: Rob ter Heine <[email protected]>
To: [email protected]; [email protected]
Sent: Friday, February 6, 2009 5:54:33 AM
Subject: Betr.: Re: [NMusers] simulation question
Dear Ethan,
Have you tried removing the $ESTIMATION step and replace all values with your
final estimations? What does your result file say after termination?
Sincerely
Rob
>>> Sebastian Ueckert <[email protected]> 02/06/09 10:51 >>>
On Thu, Feb 5, 2009 at 11:18 PM, Ethan Wu <[email protected]> wrote:
> Dear users,
> I am trying to compare several specific PK/PD study designs by:
> -- run 200 simulations with the final model (develope from original
> dataset)
> -- fit the final model to the 200 simulated dataset
> To achieve above, I used $SIM SUBPROB=200 option
> however, nonmem would completely stop after running
> into estimation problem at one specific simulation/estimation cycles, for
> some designs it stop even before 10th iterations.
> Is there anyway nonmem could continue go on?
> Or, does someone know alternative way to achieve the goal?
> thanks
>
>
>
Do you bootstrap the individual patients X times and then fit the model
to these new datasets or do you simulate datasets based on the final
population characteristics you already established? Both options are
also automated in the PDX-MC-PEM program and I think also in SADAPT.
Best
Serge
Quoted reply history
________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of Sebastian Ueckert
Sent: Friday, February 06, 2009 1:51 AM
To: Ethan Wu
Cc: [email protected]
Subject: Re: [NMusers] simulation question
Dear Ethan,
the simplest solution to solve your problem would be to use the
bootstrap command of PSN ( http://psn.sourceforge.net/). With PSN
installed you would simply do:
bootstrap final_model.mod -samples=200
PSN would take care of unsuccessful runs and provide a nice summary of
the individual estimates.
Best regards
Sebastian
On Thu, Feb 5, 2009 at 11:18 PM, Ethan Wu <[email protected]> wrote:
Dear users,
I am trying to compare several specific PK/PD study designs by:
-- run 200 simulations with the final model (develope from original
dataset)
-- fit the final model to the 200 simulated dataset
To achieve above, I used $SIM SUBPROB=200 option
however, nonmem would completely stop after running into estimation
problem at one specific simulation/estimation cycles, for some designs
it stop even before 10th iterations.
Is there anyway nonmem could continue go on?
Or, does someone know alternative way to achieve the goal?
thanks
--
The information contained in this email message may
contain confidential or legally privileged information and is intended solely
for the use of the named recipient(s). No confidentiality or privilege is
waived or lost by any transmission error. If the reader of this message is
not the intended recipient, please immediately delete the e-mail and all
copies of it from your system, destroy any hard copies of it and notify the
sender either by telephone or return e-mail. Any direct or indirect use,
disclosure, distribution, printing, or copying of any part of this message is
prohibited. Any views expressed in this message are those of the individual
sender, except where the message states otherwise and the sender is
authorized to state them to be the views of XOMA.