NONMEM Users Network Archive

Hosted by Cognigen

Re: RE: Link different thetas to same omega using mu referencing

From: Emmanuel Chigutsa <echigutsa>
Date: Wed, 13 Apr 2016 17:00:50 +0000 (UTC)

I think the following code should work:
CL = EXP(MU_1+ETA(1))*(1+CLCOV)

      From: Rupert Austin <rupert.austin
 To: "'Sadler, Brian'" <Brian.Sadler
 Sent: Wednesday, April 13, 2016 11:44 AM
 Subject: RE: [NMusers] RE: Link different thetas to same omega using mu re=
#yiv7090570546 #yiv7090570546 -- _filtered #yiv7090570546 {panose-1:2 4 5 3=
 5 4 6 3 2 4;} _filtered #yiv7090570546 {font-family:Calibri;panose-1:2 15 =
5 2 2 2 4 3 2 4;} _filtered #yiv7090570546 {font-family:Tahoma;panose-1:2 1=
1 6 4 3 5 4 4 2 4;}#yiv7090570546 #yiv7090570546 p.yiv7090570546MsoNormal, =
#yiv7090570546 li.yiv7090570546MsoNormal, #yiv7090570546 div.yiv7090570546M=
soNormal {margin:0cm;margin-bottom:.0001pt;font-size:11.0pt;}#yiv7090570546=
 a:link, #yiv7090570546 span.yiv7090570546MsoHyperlink {color:#0563C1;text-=
decoration:underline;}#yiv7090570546 a:visited, #yiv7090570546 span.yiv7090=
570546MsoHyperlinkFollowed {color:#954F72;text-decoration:underline;}#yiv70=
90570546 p.yiv7090570546MsoAcetate, #yiv7090570546 li.yiv7090570546MsoAceta=
te, #yiv7090570546 div.yiv7090570546MsoAcetate {margin:0cm;margin-bottom:.0=
001pt;font-size:8.0pt;}#yiv7090570546 span.yiv7090570546BalloonTextChar {}#=
yiv7090570546 span.yiv7090570546EmailStyle19 {color:windowtext;}#yiv7090570=
546 span.yiv7090570546EmailStyle20 {color:#1F497D;}#yiv7090570546 span.yiv7=
090570546EmailStyle21 {color:#1F497D;}#yiv7090570546 span.yiv7090570546Emai=
lStyle23 {color:#1F497D;}#yiv7090570546 .yiv7090570546MsoChpDefault {font-s=
ize:10.0pt;} _filtered #yiv7090570546 {margin:70.85pt 70.85pt 70.85pt 70.85=
pt;}#yiv7090570546 div.yiv7090570546WordSection1 {}#yiv7090570546 I=
m not sure that Brian’s suggestion is acceptable due to the foll=
owing constraint I quote from one of the NONMEM manuals:  “Tim=
e dependent covariates cannot be part of the MU_ equation”  If=
 CAT is time-varying (as suggested by Jacob) then “MU_1 = CAT1*TH=
ETA(1) + CAT2*THETA(2) + CAT3*THETA(3)” includes time dependent cov=
ariates. Rather worryingly, I don’t think NONMEM will trap this pro=
blem since I have fallen foul of it with no resulting NONMEM warnings, and =
I have no idea what the consequences are for the resulting NONMEM run and r=
esults.  Rupert  Rupert Austin, PhDSenior ScientistBAST Inc Limit=
edLoughborough University Science and Enterprise ParkCharnwood WingHolywell=
 ParkAshby RoadLoughborough, LE11 3AQ, UKTel: +44 (0)1509 222908  From=
: owner-nmusers
alf Of Sadler, Brian
Sent: 13 April 2016 13:26
To: nmusers
Subject: [NMusers] RE: Link different thetas to same omega using mu referen=
cing  Hi Jacob,  I suggest:CAT1=0CAT2=0CAT3=0IF(CAT.EQ.1) C=
AT1=1IF(CAT.EQ.2) CAT2=1IF(CAT.EQ.3) CAT3=1MU_1 = CAT1*THETA(1) + C=
AT2*THETA(2) + CAT3*THETA(3)CL = EXP(MU_1+ETA(1))  That way you have=
 only one OMEGA. This could be more flexible when expanding OMEGA for addit=
ional IIV in block format.  Cheers… Brian    From: o=
 Of Michael Fossler
Sent: Wednesday, April 13, 2016 7:56 AM
To: Leander, Jacob; nmusers
Subject: [NMusers] RE: Link different thetas to same omega using mu referen=
cing  This looks like it would work – it seems to obey the rul=
es around mu-referencing. You can simplify the $OMEGA by using SAME(3) =
 MikeMichael J. Fossler, Pharm. D., Ph. D., F.C.P.VP, Quantitative=
 SciencesTrevena, Incmfossler
ell: 610-329-6636  From: owner-nmusers
Sent: Wednesday, April 13, 2016 7:01 AM
To: nmusers
Subject: [NMusers] Link different thetas to same omega using mu referencing=
  Hi  Consider the case where we have a categorical (potentially =
time-varying) covariate named CAT that can take several values (e.g. 0,1,2,=
3).Let’s say I want to model CL depending on this categorical covar=
iate (where parameters are estimated on log-scale). A simple solution to th=
is would be  IF(CAT.EQ.1) TVCL = THETA(1)IF(CAT.EQ.2) TVCL = THETA=
(2)IF(CAT.EQ.3) TVCL = THETA(3)  I let CL be described by a log-norm=
al distribution with the same omega element for all CAT values.  CL =
= EXP(TVCL + ETA(1))  With an omega statement:  $OMEGA 0.1 =
I now want to switch to the mu referencing framework in NONMEM. Each the=
ta needs to be reference to a MU value, and to force it to be described by =
the same eta I apply the following approach.  MU_1 = THETA(1)MU_2 =
= THETA(2)MU_3 = THETA(3)  IF(CAT.EQ.1) CL = EXP(MU_1 + ETA(1))I=
F(CAT.EQ.2) CL = EXP(MU_2 + ETA(2))IF(CAT.EQ.3) CL = EXP(MU_3 + ETA(3))=
  With an omega statement:  $OMEGA BLOCK(1) 0.1$OMEGA BLOCK(1) SA=
ME$OMEGA BLOCK(1) SAME  where I force the omega element corresponding =
to ETA1, ETA2 and ETA3  to be the same.  Is this a valid approach=
 to obtain what I need? Or is there a simpler way in the mu-referencing fra=
mework to link different thetas to the same omega element?  All the be=
st,Jacob  Confidentiality Notice: This message is private and may con=
tain confidential and proprietary information. If you have received this me=
ssage in error, please notify us and remove it from your system and note th=
at you must not copy, distribute or take any action in reliance on it. Any =
unauthorized use or disclosure of the contents of this message is not permi=
tted and may be unlawful.  
Notice: This e-mail message, together with any attachments, contains inform=
ation of Trevena, Inc., 1018 West 8th Avenue, King of Prussia, PA 19406, US=
A. This information may be confidential, proprietary, copyrighted and/or le=
gally privileged.
It is intended solely for use by the individual or entity named on this mes=
sage. If you are not the intended recipient, and have received this message=
 in error, please notify us immediately and delete it and any attachments f=
rom your system.

ICON plc made the following annotations.
This e-mail transmission may contain confidential or legally privileged inf=
ormation 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 notifie=
d that any disclosure, copying, distribution, or reliance upon the contents=
 of this e-mail is strictly prohibited. If you have received this e-mail tr=
ansmission in error, please reply to the sender, so that ICON plc can arran=
ge for proper delivery, and then please delete the message.

Thank You,

ICON plc
South County Business Park
Dublin 18
Registered number: 145835


Received on Wed Apr 13 2016 - 13:00:50 EDT

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to:

Once subscribed, you may contribute to the discussion by emailing: