NONMEM Users Network Archive

Hosted by Cognigen

Re: Use AND to ignore rows for input dataset

From: Mark Tepeck <mark.tepeck>
Date: Tue, 17 Dec 2019 20:20:54 -0800

Hi Bob,

Thanks for providing the solution applicable to some cases. Meanwhile,
I would like to share my thoughts on this issue. My sole purpose is to
make NONMEM even better.

1. Use of OR alone may be subject to a very limited utility. Actually,
my scenario is way more complicated. E.g. I want to exclude the
year-span between (10,18) and (60,70) from the pool of (0,90). I
cannot find a solution with ACCEPT/OR for this case.

2. Implicit/default use of OR is quite counter-intuitive and could
raise the NONMEM learning curve. As we all know, AND (instead of OR)
is the widely implemented default operator, E.g., Google search and
programming languages (like R package dplyr/filter). NONMEM developers
may consider following this general practice.

3. A lot of solutions provided on this platform come from the Guide
-VIII. They are still useful but seems cumbersome from the current
perspective. While we stick to a six-year-old guide, we entered a new
era of machine learning, artificial intelligence, self-driving car,
smart phones, etc, technologies are evolving quickly to become much
easier than ever to use. Maybe, we can make NONMEM more user-friendly
as well by solving this long-standing issue. 😊 With this fixed,
IGNORE can serve as a universal filtering command.

I really appreciate your input and feel free to correct me if I am wrong.

Mark


On Mon, Dec 16, 2019 at 9:41 PM Bauer, Robert <Robert.Bauer
ote:
>
> Sebastien:
>
> It is not clear in the guide, but only parenthesized IGNORE and ACCEPT li=
sts, such as IGNORE=(list) and ACCEPT=(list), are incompatible. IGNOR=
E=c, where c could be any character except space, should work with IGNORE=
=(list) or ACCEPT=(list). For example:
>
>
>
> $DATA example1.csv ACCEPT=(CMT.EQN.1) IGNORE=
>
>
>
> Also, any data record starting with first non-blank character of # will a=
lways be ignored by NMTRAN.
>
> Robert J. Bauer, Ph.D.
>
> Senior Director
>
> Pharmacometrics R&D
>
> ICON Early Phase
>
> 820 W. Diamond Avenue
>
> Suite 100
>
> Gaithersburg, MD 20878
>
> Office: (215) 616-6428
>
> Mobile: (925) 286-0769
>
> Robert.Bauer
>
> www.iconplc.com
>
>
>
> From: Sebastien Bihorel <Sebastien.Bihorel
> Sent: Monday, December 16, 2019 5:02 PM
> To: Bauer, Robert <Robert.Bauer
.com>
> Subject: Re: [NMusers] Use AND to ignore rows for input dataset
>
>
>
> Since nowadays most datasets include a header line, IGNORE=
ing along those lines) is almost always used in control streams. Since ACCE=
PT and IGNORE cannot be used in the same $DATA record, the suggested techni=
que is seldom usable in my experience.
>
> ________________________________
>
> From: owner-nmusers
lf of Bauer, Robert <Robert.Bauer
> Sent: Monday, December 16, 2019 17:39
> To: Nmusers <nmusers
> Subject: RE: [NMusers] Use AND to ignore rows for input dataset
>
>
>
> Guide VIII suggests the following:
>
>
>
> Suppose it is desired that records be dropped that satisfy the logical ".=
AND." of several conditions. This can be implemented by using an ACCEPT lis=
t with the negations of the conditions. For example, suppose that records t=
o be ignored are those having GEN=1 .AND. AGE > 60. This may be done as f=
ollows:
>
> ACCEPT=(GEN.NE.1,AGE.LE.60)
>
>
>
> Robert J. Bauer, Ph.D.
>
> Senior Director
>
> Pharmacometrics R&D
>
> ICON Early Phase
>
> 820 W. Diamond Avenue
>
> Suite 100
>
> Gaithersburg, MD 20878
>
> Office: (215) 616-6428
>
> Mobile: (925) 286-0769
>
> Robert.Bauer
>
> www.iconplc.com
>
>
>
> From: owner-nmusers
lf Of Sebastien Bihorel
> Sent: Monday, December 16, 2019 12:39 PM
> To: Mark Tepeck <mark.tepeck
> Subject: Re: [NMusers] Use AND to ignore rows for input dataset
>
>
>
> I would second that feature request!
>
> ________________________________
>
> From: owner-nmusers
lf of Mark Tepeck <mark.tepeck
> Sent: Monday, December 16, 2019 15:16
> To: Nmusers <nmusers
> Subject: [NMusers] Use AND to ignore rows for input dataset
>
>
>
> Hi Colleagues,
>
>
>
> Although there are many workarounds to manage the input datasets for NONM=
EM, use of IGNORE would be a very handy solution. However, I am surprised =
to see that AND is not supported by NONMEM. e.g. IGNORE=(CMT.EQ.1, AND C=
MT.EQ.2 ) would throw an error. In contrast, NONMEM use OR for IGNORE=(C=
MT.EQ.1, CMT.EQ.2 ) by default, which is quite different for the common pra=
ctice of software.
>
>
>
> I am wondering if the AND feature has been or will be implemented in the =
last/future NONMEM version.
>
>
>
> Thank you,
>
>
>
> Mark
>
>
>
> ICON plc made the following annotations.
> -------------------------------------------------------------------------=
-----
> This e-mail transmission may contain confidential or legally privileged i=
nformation that is intended only for the individual or entity named in the =
e-mail address. If you are not the intended recipient, you are hereby notif=
ied that any disclosure, copying, distribution, or reliance upon the conten=
ts of this e-mail is strictly prohibited. If you have received this e-mail =
transmission in error, please reply to the sender, so that ICON plc can arr=
ange for proper delivery, and then please delete the message.
>
> Thank You,
>
> ICON plc
> South County Business Park
> Leopardstown
> Dublin 18
> Ireland
> Registered number: 145835
Received on Tue Dec 17 2019 - 23:20:54 EST

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to: nmusers-request@iconplc.com.

Once subscribed, you may contribute to the discussion by emailing: nmusers@globomaxnm.com.