From: Elassaiss - Schaap, J. - <*jeroen.elassaiss*>

Date: Thu, 24 Jun 2010 07:25:08 +0200

Dear Hauke,

Nonmem cannot handle random variables defined in nested conditions. See =

the help entry on abbreviated code:

"Random variables cannot be defined within nested conditionals, i.e.,

within a conditional structure beginning with an IF and containing

another IF. The use of ELSEIF ... THEN implies a nested conditional."

The usual way of handling this is to define a dummy variable to set the =

condition, but you do not even need this in your case as Brendan and =

Katya pointed out.

Moreover in this case, I would assign a different sigma to PK and PD. =

Your model actually states the belief that the variability in PK and PD =

are 100% correlated (unless they are always measured at different =

times). If you have background support for such a model it is fine, but =

in general I would advice to associate EPS(1) with THETA(8) for PK and =

EPS(2) with THETA(9) for PD.

You may even skip zero testing by request (ONLYOBS) after $ERROR if you =

always have positive predictions of C1 and C3 during observation records =

(e.g., a PK model without lagtime), see nmhelp on $ERROR.

Hope this helps,

Jeroen

Modeling & Simulation Expert

Pharmacokinetics, Pharmacodynamics & Pharmacometrics (P3) - DMPK

MSD

PO Box 20 - AP1112

5340 BH Oss

The Netherlands

jeroen.elassaiss

T: +31 (0)412 66 9320

F: +31 (0)412 66 2506

www.msd.com

________________________________

From: owner-nmusers

On Behalf Of Hauke Rühs

Sent: Wednesday, 23 June, 2010 14:54

To: nmusers

Subject: [NMusers] simultaneous PK/PD

Dear NMusers,

I want to combine my sequential PK and PD model to one simultaneous =

model. In both models I used an exponential error model =

(log-transformation). When I write the model as follows, NONMEM gives me =

this error message:

326 RANDOM VARIABLE IS DEFINED IN A NESTED IF STRUCTURE.

Leaving out the condition C.GT.0 will also lead to errors. Is there =

another way to write this?

Thanks

Hauke

$ERROR

C1=(A(1)/V1)

C3=A(3)

IPRE=0

IF (M1H2.EQ.1) THEN

IF (C1.GT.0) IPRE = LOG(C1)

IRES = DV-IPRE

W = EPS(1)*THETA(8)

Y = IPRE+W

ENDIF

IF (M1H2.EQ.2) THEN

IF (C3.GT.0) IPRE = LOG(C3)

IRES = DV-IPRE

W = EPS(1)*THETA(9)

Y = IPRE+W

ENDIF

DEL = 0

IF(W.EQ.0) DEL = 1

IWRE = IRES/(W+DEL)

$SIGMA

1 FIX

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

Hauke Rühs

Apotheker

Pharmazeutisches Institut

- Klinische Pharmazie -

An der Immenburg 4

53121 Bonn

Tel: + 49-(0)228 73-5781

Fax: + 49-(0)228 73-9757

www.klinische-pharmazie.info <http://www.klinische-pharmazie.info/>

This message and any attachments are solely for the intended recipient. =

If you are not the intended recipient, disclosure, copying, use or =

distribution of the information included in this message is prohibited =

--- Please immediately and permanently delete.

Received on Thu Jun 24 2010 - 01:25:08 EDT

Date: Thu, 24 Jun 2010 07:25:08 +0200

Dear Hauke,

Nonmem cannot handle random variables defined in nested conditions. See =

the help entry on abbreviated code:

"Random variables cannot be defined within nested conditionals, i.e.,

within a conditional structure beginning with an IF and containing

another IF. The use of ELSEIF ... THEN implies a nested conditional."

The usual way of handling this is to define a dummy variable to set the =

condition, but you do not even need this in your case as Brendan and =

Katya pointed out.

Moreover in this case, I would assign a different sigma to PK and PD. =

Your model actually states the belief that the variability in PK and PD =

are 100% correlated (unless they are always measured at different =

times). If you have background support for such a model it is fine, but =

in general I would advice to associate EPS(1) with THETA(8) for PK and =

EPS(2) with THETA(9) for PD.

You may even skip zero testing by request (ONLYOBS) after $ERROR if you =

always have positive predictions of C1 and C3 during observation records =

(e.g., a PK model without lagtime), see nmhelp on $ERROR.

Hope this helps,

Jeroen

Modeling & Simulation Expert

Pharmacokinetics, Pharmacodynamics & Pharmacometrics (P3) - DMPK

MSD

PO Box 20 - AP1112

5340 BH Oss

The Netherlands

jeroen.elassaiss

T: +31 (0)412 66 9320

F: +31 (0)412 66 2506

www.msd.com

________________________________

From: owner-nmusers

On Behalf Of Hauke Rühs

Sent: Wednesday, 23 June, 2010 14:54

To: nmusers

Subject: [NMusers] simultaneous PK/PD

Dear NMusers,

I want to combine my sequential PK and PD model to one simultaneous =

model. In both models I used an exponential error model =

(log-transformation). When I write the model as follows, NONMEM gives me =

this error message:

326 RANDOM VARIABLE IS DEFINED IN A NESTED IF STRUCTURE.

Leaving out the condition C.GT.0 will also lead to errors. Is there =

another way to write this?

Thanks

Hauke

$ERROR

C1=(A(1)/V1)

C3=A(3)

IPRE=0

IF (M1H2.EQ.1) THEN

IF (C1.GT.0) IPRE = LOG(C1)

IRES = DV-IPRE

W = EPS(1)*THETA(8)

Y = IPRE+W

ENDIF

IF (M1H2.EQ.2) THEN

IF (C3.GT.0) IPRE = LOG(C3)

IRES = DV-IPRE

W = EPS(1)*THETA(9)

Y = IPRE+W

ENDIF

DEL = 0

IF(W.EQ.0) DEL = 1

IWRE = IRES/(W+DEL)

$SIGMA

1 FIX

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

Hauke Rühs

Apotheker

Pharmazeutisches Institut

- Klinische Pharmazie -

An der Immenburg 4

53121 Bonn

Tel: + 49-(0)228 73-5781

Fax: + 49-(0)228 73-9757

www.klinische-pharmazie.info <http://www.klinische-pharmazie.info/>

This message and any attachments are solely for the intended recipient. =

If you are not the intended recipient, disclosure, copying, use or =

distribution of the information included in this message is prohibited =

--- Please immediately and permanently delete.

Received on Thu Jun 24 2010 - 01:25:08 EDT