NONMEM Users Network Archive

Hosted by Cognigen

RE: [NMusers] Incomplete circles drawn in VPC plotted using R scripts generated by PsN

From: <E.Olofsen_at_lumc.nl>
Date: Fri, 20 Nov 2015 12:55:52 +0000

I'm not sure if this is related, but I always use the "useDingbats=F" opt=
ion when creating pdfs in R. =

Erik =
________________________________________ =
From: owner-nmusers_at_globomaxnm.com [owner-nmusers_at_globomaxnm.com] on behalf=
 of Standing Joseph (GREAT ORMOND STREET HOSPITAL FOR CHILDREN NHS FOUNDATI=
ON TRUST) [joseph.standing_at_nhs.net] =
Sent: Friday, November 20, 2015 1:23 PM =
To: Kajsa Harling; HUI, Ka Ho; nmusers_at_globomaxnm.com =
Subject: RE: [NMusers] Incomplete circles drawn in VPC plotted using R scri=
pts generated by PsN =

I think it is a problem with R making pdfs. Workaround: try using tiff() =
or png() instead - then you can save the image as a pdf. =
Joe =

________________________________________ =
From: owner-nmusers_at_globomaxnm.com [owner-nmusers_at_globomaxnm.com] On Behalf=
 Of Kajsa Harling [kajsa.harling_at_farmbio.uu.se] =
Sent: 20 November 2015 10:33 =
To: HUI, Ka Ho; nmusers_at_globomaxnm.com =
Subject: Re: [NMusers] Incomplete circles drawn in VPC plotted using R scri=
pts generated by PsN =

Dear Matthew, =

I suggest you consult the xpose.VPC documentation ( ?xpose.VPC() ) for inst=
ructions on how to change how values are marked in the plot. The PsN -rplot=
s script uses the default settings. =

Best regards, =
Kajsa =

On 11/19/2015 03:56 PM, HUI, Ka Ho wrote: =
Dear all, =

I was trying to generate VPC using the R scripts generated by PsN with the =
following command line: =

vpc =
-samples=200 =
-stratify_on=… =
-rplots=1 =
-idv=TIME =
-auto_bin=unique =
-predcorr =
-varcorr =
-dir=vpc_FINAL_MODEL =
FINAL_MODEL_vpc.mod =

The R script generated looks like this: =

#START OF AUTO-GENERATED PREAMBLE, WILL BE OVERWRITTEN WHEN THIS FILE IS US=
ED AS A TEMPLATE =
#Created 2015-11-19 at 22:05 =

rplots.level <- 1 =
xpose.runno <- '1_vpc' =
toolname <- 'vpc' =
pdf.filename <- paste0('PsN_',toolname,'_plots.pdf') =
pdf.title <- 'vpc diagnostic plots run 1_vpc' =
working.directory<-'C:/…' =
raw.results.file <- 'raw_results_FINAL_MODEL1_vpc.csv' =
model.directory<-'C:/…' =
model.filename<-'FINAL_MODEL1_vpc.mod' =
subset.variable<-NULL =
mod.suffix <- '.mod' =
mod.prefix <- 'FINAL_MODEL' =
tab.suffix <- '' =
tool.results.file <- 'vpc_results.csv' =
theta.labels <- c('T1','T2','T3','T4','T5','T6') =
theta.fixed <- c(FALSE,FALSE,TRUE,FALSE,FALSE,FALSE) =
omega.labels <- c('O1','O2') =
omega.fixed <- c(FALSE,FALSE) =
sigma.labels <- c('S1') =
sigma.fixed <- c(FALSE) =
n.eta <- 2 =
n.eps <- 1 =

vpctab <- 'vpctab' =
have.loq.data <- FALSE =
have.censored <- FALSE =
is.categorical <- FALSE =
is.tte <- FALSE =
dv <- 'DV' =
idv <- 'TIME2' =

setwd(working.directory) =

###########################################################################=
# =
#END OF AUTO-GENERATED PREAMBLE =
#WHEN THIS FILE IS USED AS A TEMPLATE THIS LINE MUST LOOK EXACTLY LIKE THIS=



library(xpose4) =

pdf(file=pdf.filename,width=10,height=7,title=pdf.title) =

done <- FALSE =
if (is.tte){ =
     #data is in the model directory, go there to read input =
        setwd(model.directory) =
        xpdb <- xpose.data(xpose.runno) =
        plots <- kaplan.plot(object=xpdb,VPC=T) =
        #go back to vpc directory =
        setwd(working.directory) =
        done <- TRUE =
} =

if (is.categorical & (!done)){ =
    plots<-xpose.VPC.categorical(vpc.info=tool.results.file,vpctab=vpct=
ab) =
        done <- TRUE =
} =

if ((have.loq.data | have.censored) & (!done) ){ =
    plots<-xpose.VPC.both(vpc.info=tool.results.file,vpctab=vpctab) =
        done <- TRUE =
} =

if (!done){ =
        plots<-xpose.VPC(vpc.info=tool.results.file,vpctab=vpctab) =
        done <- TRUE =
} =

print(plots) =

dev.off() =

However the “circles” (indicating raw Cp data) are incomplete and take =
the shape of a “crescent moon” instead. When run with the –lloq optio=
n, the 1st stratification has complete circles but the remaining 2 also has=
 the crescent-moon circles instead. Other parts of the plots (Shades and pl=
ots of CI) look normal. =
Can anyone suggest any possible cause and solution? =

Thanks and regards, =
Matthew Hui =


-- =
----------------------------------------------------------------- =
Kajsa Harling, PhD =
System Developer =
Department of Pharmaceutical Biosciences =
Uppsala University =

Kajsa.Harling_at_farmbio.uu.se<mailto:Kajsa.Harling_at_farmbio.uu.se> =
+46-(0)18-471 4308 =

http://www.farmbio.uu.se/research/researchgroups/pharmacometrics/ =
----------------------------------------------------------------- =



***************************************************************************=
***************************************** =

This message may contain confidential information. If you are not the inten=
ded recipient please inform the =
sender that you have received the message in error before deleting it. =
Please do not disclose, copy or distribute information in this e-mail or ta=
ke any action in reliance on its contents: =
to do so is strictly prohibited and may be unlawful. =

Thank you for your co-operation. =

NHSmail is the secure email and directory service available for all NHS sta=
ff in England and Scotland =
NHSmail is approved for exchanging patient data and other sensitive informa=
tion with NHSmail and GSi recipients =
NHSmail provides an email address for your career in the NHS and can be acc=
essed anywhere =

***************************************************************************=
***************************************** =
Received on Fri Nov 20 2015 - 07:55:52 EST

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to: nmusers-request_at_iconplc.com. Once subscribed, you may contribute to the discussion by emailing: nmusers@globomaxnm.com.