- On 2 Jun 2004 at 12:16:06, "anita" (anita.-at-.pharmacy.bg.ac.yu) sent the message
Dear all,

Back to the Top

We would kindly ask those experienced in population PK to help us to

choose

the most appropriate (and the most friendly user) software for

population PK

modeling. We would use it more for scientific researches than for

practice.

We have some modest experience with NONMEM, and so far it seems very

complicated for many different reasons (before all, it operates under

DOS

and FORTRAN). Is there any alternative for NONMEM, which would be

"easier"

for use, but would offer the same spectra of PK estimations.

Thanks.

Anita - On 3 Jun 2004 at 12:01:26, "Ravi Kuppuraj" (rkuppuraj.at.innaphase.com) sent the message
Anita,

Back to the Top

Kinetica has probably the easiest-to-use Population PK module out

there. This is an integral part of Kinetica, which can also be used to

perform NCA, and conventional fitting/compartmental modeling as well.

You can download a demo version of Kinetica from

http://www.innaphase.com/support_downloads_kinetica.html

Please email me for a password to unzip the demo version. The software

is identical to the commercial package, except that without a license

you will not be able to save or print. NCA,compartmental and population

analyses are possible using the pre-built templates and models, as well

as a graphic designer to build your own models.

Ravi Kuppuraj - On 3 Jun 2004 at 11:37:24, Roger Jelliffe (jelliffe.aaa.usc.edu) sent the message
Dear Anita:

Back to the Top

You might consider using the USC*PACK collection of programs,

which have the nonparametric population software NPAG for both small

models with analytic solutions to the differential equations and larger

and nonlinear models which use differential equation solvers, They

currently run in DOS. If you are interested in software that is

mathematically consistent (the more subjects studied, the closer the

results are to the truth), and the most statistically efficient of what

is available now (the most precise parameter estimates), you might

consider using this software. You can get more information from our web

site www.lapk.org, where you can click on new advances in population

modeling, and also on many teaching topics.

Very best regards,

Roger Jelliffe

Roger W. Jelliffe, M.D. Professor of Medicine,

Division of Geriatric Medicine,

Laboratory of Applied Pharmacokinetics,

USC Keck School of Medicine

2250 Alcazar St, Los Angeles CA 90033, USA

Phone (323)442-1300, fax (323)442-1302, email= jelliffe.at.usc.edu

Our web site= http://www.lapk.org - On 3 Jun 2004 at 23:49:10, Wojciech Jawien (mfjawien.aaa.cyf-kr.edu.pl) sent the message
Dear Anita,

Back to the Top

In my opinion, there are at least four different groups of programs

which might be used for pop PK.

The NONMEM falls to the group in which you can also find WinNonMix from

Pharsight (www.pharsight.com). If you are familiar with WinNonlin, you

will find user interface very similar. It is certainly easier to use.

There is also a procedure PROC NLMIXED in a SAS/STAT which does the

same job, namely using optimization algorithms looks for a maximum

likelihood of a normally distributed parameters, the standard

estimation technique with a graeat number of sofisticated tricks which

make computations feasible.

Let's go back to the NONMEM for a while. First, you may use it also

under operating system different from DOS, of which Linux might be of a

special interest, because it includes ready to use Fortran compiler,

everything (besides NONMEM itself) for free. Second, there exist some

software which makes life with NONMEM under DOS/Windows much easier.

The one I know is Visual NONMEM from R.D.P.P., France

(http://members.aol.com/rdppweb/vnm.htm). You might also had heard

about 'Wings for Nonmem'.

Another point is that NONMEM is complicated because the idea of the

population PK/PD is complicated, and the easiest user interface doesn't

relax you from a need to deeply understand important ideas.

Closely related, yet another group consists of Kinetica, mentioned by

Ravi. It also looks for a maximum likelihood, but instead of a general

purpose optimization tools uses a special version of Dempster's

Expectation Maximization (EM) algorithm. In fact Kinetica's population

module appears to be a descendant of a PPharm program.

Relaxing the assumption on normality allows for two other groups. The

programs from Dr. Jelliffe's team, USC, Los Angeles do the

non-parametric (in a statistical sense) max-likelihood estimation.

These are NPEM2 which employs non-parametric version of EM, and NPAG

which is not EM-like (www.lapk.org), and appeas to be more user

friendly, however it is still under development.

The fourth approach is a Monte-Carlo Markov Chain. There is a module

called PKBugs which can be used with a general-purpose software

WinBugs. Here, in the Bayesian framework (but Nonmem also works within

this framework) one can not only find mean and SD of populaton model

parametrs, as well as their individual expected values, but it allows

to determine estimated full probability distribution of them. This

approach is parametric in that sense, it requires assumptions on

distributions of parameters, but one need not to be bound with a normal

distribution. PKBugs allows only for several simple models, however it

is possible to define others with a WinBugs. The philosophy of working

with PKBugs/WinBugs is quite different from all other approaches (but

data are expected to be in Nonmem-like format). User interface itself

is excelent (in the Windows environment). See

http://www.med.ic.ac.uk/divisions/60/pkbugs_web/home.html.

Well, if you had any idea, now you might lose it...

I recommend Dr. Aarons paper: Clin.Pharmacokin. 36:255-264(1999).

Maybe someone on the list knows a newer review on pop PK/PD software.

You may also wish to consult eg.: Lunn D. et al.:

J.Pharmacokin.Pharmacodyn. 29:271-307(2002).

Best regards

Wojciech Jawien

Jagiellonian Univ.

Faculty of Pharmacy

Krakow, Poland - On 4 Jun 2004 at 09:14:07, "Peter Moate" (moate.-a-.vet.upenn.edu) sent the message
Dear Anita and Wojciech,

Back to the Top

WinSAAM also is very good, inexpensive and simple to use for

population modeling of compartmental systems. A description of the population

modeling tool in WinSAAM can be found in: Stefanovski D, Moate PJ, Boston RC.

WinSAAM: A windows-based compartmental modeling system. Metabolism.

2003;52(9):1153-1166.

Also, information about WinSAAM can be found at: www.winsaam.com

Regards,

Peter Moate - On 6 Jun 2004 at 11:11:02, "NM.at.drudevo.com" (nm.aaa.drudevo.com) sent the message
Dear all,

Back to the Top

If you are looking for such a software, take a look at www.drudevo.com.

NONMEM might sometimes be difficult, but MME(previous NMI) allows you

to do

modelling just as easy as in other software using NONMEM as an engine.

It

will also do automatic model finding by a couple of mouse clicks without

knowing NONMEM and/or its datastructure.

Want to post a follow-up message on this topic? If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Software for population PK" as the subject

PharmPK Discussion List Archive Index page

Copyright 1995-2010 David W. A. Bourne (david@boomer.org)