inst/poppk_report_latex_doNotUse/Setup/figures.R
In AstraZeneca/pmworkbench: Respoitory of templates for pharmacometric projects
###################################################
# figures.R
#
# Author: Helena Edlund
# Created on: 2017-07-26
# Last modified on:
# Purpose: Produce all project related figures contained in the report
# Dependencies: environmentPrep.R
###################################################
# -------------------------------------------------
# Goodness of fit for base model
# -------------------------------------------------
grid.arrange(
ggplot(baseModelConcData, aes(y=DV, x=PRED)) +
geom_hline(aes(yintercept=0), linetype="dashed", colour="darkgray") +
geom_abline(slope=1, intercept=0) +
geom_point(shape=1, size = 3, alpha=0.4) +
geom_smooth(method="loess") +
coord_cartesian(xlim=c(minConc,maxConc), ylim=c(minConc,maxConc)) +
labs(y="ln(Observed)", x="ln(Population prediction)") +
settings,
ggplot(baseModelConcData, aes(y=DV, x=IPRED)) +
geom_hline(aes(yintercept=0), linetype="dashed", colour="darkgray") +
geom_abline(slope=1, intercept=0) +
geom_point(shape=1, size = 3, alpha=0.4) +
geom_smooth(method="loess") +
coord_cartesian(xlim=c(minConc,maxConc), ylim=c(minConc,maxConc)) +
labs(y="Observed", x="Individual prediction") +
settings,
ggplot(baseModelConcData, aes(y=CWRES, x=TAPD)) +
geom_hline(aes(yintercept=0), color="darkgray") +
geom_point(shape=1, size = 3, alpha=0.4) +
geom_smooth(method="loess") +
coord_cartesian(ylim=c(-maxCWRES,maxCWRES)) +
scale_y_continuous(breaks=seq(from=-20, to=20, by=2)) +
labs(y="CWRES", x="Time after dose") +
settings,
ggplot(baseModelConcData, aes(y=CWRES, x=PRED)) +
geom_point(shape=1, size = 3, alpha=0.4) +
geom_hline(aes(yintercept=0)) +
geom_smooth(method="loess") +
coord_cartesian(ylim=c(-maxCWRES,maxCWRES)) +
scale_y_continuous(breaks=seq(from=-20, to=20, by=2)) +
labs(y="CWRES", x="ln(Population prediction)") +
settings,
nrow=2)
# -------------------------------------------------
# ETA versus covariates for base model
# -------------------------------------------------
baseModelBaseline <- baseModelConcData[!duplicated(baseModelConcData$ID),]
p <-
ggduo(baseModelBaseline,
c('AGE','BWT','BHT','BEGFR','BBILI','BAST'),
c('ETA3',"ETA4"),
types = list(continuous = scatterWithSmooth),
columnLabelsX = c('AGE','BWT','BHT','BEGFR','BBILI','BAST'),
columnLabelsY = c("Random effect of CL/F",'Random effect of Vc/F'))
print(p, leftWidthProportion = 0.35)
p <-
ggduo(baseModelBaseline ,
c('SEXM','RACEL1','ETHNIC','HEALTHGPL','ECOG','DOSE','HEPATICL','ARA'),
c('ETA3',"ETA4"),
types = list(comboVertical = myBoxplot),
columnLabelsX = c('SEX','RACE','ETHNIC','HEALTHGP','ECOG','DOSE','HEPATIC','ARA'),
columnLabelsY = c("Random effect of CL/F",'Random effect of Vc/F'))
print(p, leftWidthProportion = 0.35, bottomHeightProportion = 0.65)
# -------------------------------------------------
# Parameter distributions for base model
# -------------------------------------------------
grid.arrange(
ggplot(baseModelBaseline) +
geom_histogram(aes(x=CL), fill="#737373", colour="#525252", size = 1,bins = 30) +
labs(x='EBE of CL/F', y='Count') + settings,
ggplot(baseModelBaseline) +
geom_histogram(aes(ETA3), fill="#737373", colour="#525252", size = 1,bins = 30) +
# stat_function(fun = dnorm, args = list(sd = sqrt(0.237)), size = 1, linetype = 2) +
labs(x='Random effect of CL/F', y='Count') + settings,
ggplot(baseModelBaseline) +
geom_histogram(aes(x=V2), fill="#737373", colour="#525252", size = 1,bins = 30) +
labs(x='EBE of Vc/F', y='Count') + settings,
ggplot(baseModelBaseline) +
geom_histogram(aes(ETA4), fill="#737373", colour="#525252", size = 1,bins = 30) +
# stat_function(fun = dnorm, args = list(sd = sqrt(0.237)), size = 1, linetype = 2) +
labs(x='Random effect of Vc/F', y='Count') + settings,
nrow=2)
# -------------------------------------------------
# Individual fits for base model
# -------------------------------------------------
## Select 10 best and 10 worst fit based on CWRES.
largestCWRES <-
ddply(baseModelConcData[baseModelConcData$ID !=446,], # ID 446 has only 1 conc
.(ID), summarize,
maxCWRES = max(abs(CWRES), na.rm=T))
largestCWRES <- largestCWRES[order(largestCWRES$maxCWRES),]
best <- head(largestCWRES, n = 12)$ID
worst <- tail(largestCWRES, n = 12)$ID
bestConcData <- baseModelConcData[baseModelConcData$ID %in% best, ]
worstConcData <- baseModelConcData[baseModelConcData$ID %in% worst, ]
# Remove one sparse occasion on a different dosing regimen.
# 2 rich profiles at 200 mg BID ans since plot facets on regimen the sparse
# gets a separate plot area. Does not add much to the figure so excluded.
worstConcData <-
worstConcData[!(worstConcData$ID==337 & worstConcData$Regimen=="100 mg BID"),]
ggplot(data=bestConcData[!is.na(bestConcData$OCC),],
aes(x=TAPD, y=DV, colour=factor(OCC), group=ID.OCC)) +
# observed, rich
geom_point(aes(shape=factor(BLQ)), size=2.5) +
scale_shape_manual(values=c(1,4)) + guides(shape="none") +
# population prediction, rich
geom_line(aes(x=TAPD, y=PRED, group=ID.OCC), colour="#575757",
linetype="solid", inherit.aes = F, size=1) +
# individual prediction, rich
geom_line(aes(x=TAPD, y=IPRED, colour=factor(OCC), group=ID.OCC),
linetype="longdash", size=1) +
# observed, sparse
geom_point(data=bestConcData[is.na(bestConcData$OCC),],
aes(y=DV, x=TAPD), size=2.5, shape=1, col="#575757") +
# predicted, sparse
geom_point(data=bestConcData[is.na(bestConcData$OCC),],
aes(y=IPRED, x=TAPD), size=2.5, shape=2, col="#575757") +
# lloq
geom_hline(aes(yintercept=log(1)), linetype="dashed") +
# settings
coord_cartesian(xlim=c(0,12), ylim=c(0,8)) +
scale_x_continuous(breaks=c(0,2,4,6,8,10,12)) +
scale_colour_brewer(palette="Dark2") +
guides(colour="none") +
facet_wrap(~ID+Regimen, ncol=3, labeller="label_both") +
labs(x="Time after dose (h)", y = "log(Concentration)") + theme_bw() +
settings
# -------------------------------------------------
# Forest plot for the parameter level
# -------------------------------------------------
forrest <-
covParameters[covParameters$par %in% covPar[!covPar %in% c("EGFR_CL","DOSE_CL")],
c("par","estimate","SE")]
forrest$strPar <- gsub("(.*)_","", forrest$par)
forrest$strPar <- factor(forrest$strPar, levels = c("CL", "V"), labels = c("CL/F", "Vc/F"))
forrest$par <- gsub("_.*","", forrest$par)
# estimates and intervals
forrest$estimate <- forrest$estimate+1
forrest$lCI <- forrest$estimate - 1.96*forrest$SE + 1
forrest$uCI <- forrest$estimate + 1.96*forrest$SE + 1
CL <- covParameters$estimate[covParameters$par=="CL"]
V <- covParameters$estimate[covParameters$par=="Vc"]
EGFR <- covParameters$estimate[covParameters$par=="EGFR_CL"]
DOSE <- covParameters$estimate[covParameters$par=="DOSE_CL"]
fEGFR <- data.frame(par="EGFR", estimate=1, SE=NA, strPar="CL/F",
lCI = (CL*(30/75.3)^EGFR)/CL, # min = 29.95
uCI = (CL*(170/75.3)^EGFR)/CL) # max = 280.9 but likely an outlier. next is 168.7
fDOSE <- data.frame(par="Dose", estimate=1, SE=NA, strPar="CL/F",
lCI = (CL*(15/100)^DOSE)/CL,
uCI = (CL*(400/100)^DOSE)/CL)
# allometric scaling
fBWCL <- data.frame(par="BW", estimate=1, SE=NA, strPar="CL/F",
lCI = (CL*(50/70)^0.75)/CL, # min in male patients
uCI = (CL*(140/70)^0.75)/CL) # max in male patients
fBWV <- data.frame(par="BW", estimate=1, SE=NA, strPar="Vc/F",
lCI = (V*(50/70)^1)/V,
uCI = (V*(140/70)^1)/V)
forrest <- rbind(forrest, fEGFR, fDOSE, fBWCL, fBWV)
forrest$par <-
factor(forrest$par,
levels = rev(c("ECOGHV","Dose","EGFR","ARA","SEX","RACEB","RACEO","HEP","BW")),
labels = rev(c("Health group: HV","Dose: 400-15","eGFR: 30-170",
"Acid Reducing Agent",
"Sex: Female",
"Race: Black","Race: Other",
"Hepatic: Mild/Moderate",
"Body Weight: 50-140")))
parImpact <- # printed in section below
ggplot(forrest, aes(estimate, par)) +
geom_rect(xmin = 0.8, ymin = -Inf, xmax= 1.25, ymax = Inf, fill="lightgray") +
geom_point(size=4, shape=18) +
geom_errorbarh(aes(xmax = uCI, xmin = lCI), height = 0.3) +
geom_vline(xintercept = 1, linetype = "longdash") +
theme_bw() + facet_wrap(~strPar) +
scale_x_continuous(breaks = seq(from=0, to=5, by=0.5)) +
labs(x="Change in parameter relative to population estimate", y="") +
theme(axis.title = element_text(size=12),
axis.text = element_text(size=12),
strip.text = element_text(size=10),
legend.title = element_blank(),
legend.text = element_text(size=12),
legend.margin = unit(0.1, "in"),
legend.key.size = unit(0.5, "in"),
legend.position = "bottom",
legend.box = "horizontal",
panel.margin = unit(1.2, "lines"))
# ----------------------------------------------
# Exposure by covariate
# ----------------------------------------------
# function to get numbers in boxes for graphics below
getNAUC <- function(x){
return(c(y = -200, label = length(x)))
}
getNCmax <- function(x){
return(c(y = -40, label = length(x)))
}
theme_set(theme_bw())
res <- 300
## function to divide data into 4 groups base on quartiles
# quartiles <- function(cov) {
# breaks <- quantile(cov, p=c(0.25,0.5,0.75))
# data <- data.frame(cov=cov, grouping=NA)
# data$grouping[data$cov < breaks[1]] <- "0-24.9"
# data$grouping[data$cov >= breaks[1] & data$cov < breaks[2]] <- "25-49.9"
# data$grouping[data$cov >= breaks[2] & data$cov < breaks[3]] <- "50-74.9"
# data$grouping[data$cov >= breaks[3]] <- "75-100"
#
# data$grouping <- factor(data$grouping, levels=c("0-24.9","25-49.9","50-74.9","75-100"))
# return(data$grouping)
# }
### --------- Create missing lumped categories ---------------
# Hepatic
exposureData$HEPATICL[exposureData$HEPATIC == "Normal"] <- "Normal"
exposureData$HEPATICL[exposureData$HEPATIC %in% c("Mild", "Moderate")] <- "Mild/Moderate"
exposureData$HEPATICL <- factor(exposureData$HEPATICL, levels = c("Normal","Mild/Moderate")) # NCI criteria
## Race: white, black, other
exposureData$RACEL[!is.na(exposureData$RACE) & exposureData$RACE == "White"] <- "White"
exposureData$RACEL[!is.na(exposureData$RACE) & exposureData$RACE == "Black/African American"] <- "Black/African American"
exposureData$RACEL[!is.na(exposureData$RACE) & !exposureData$RACE %in% c("White","Black/African American")] <- "Other"
exposureData$RACEL <- factor(exposureData$RACEL, levels=c("White", "Black/African American", "Other"))
# ECOGL
exposureData$ECOGL[exposureData$ECOG %in% c("HV")] <- "HV" # Healty volunteers
exposureData$ECOGL[exposureData$ECOG %in% c("0","1")] <- "ECOG: 0/1" # Healty volunteers
exposureData$ECOGL[exposureData$ECOG %in% c("2","3")] <- "ECOG: 2/3" # Healty volunteers
exposureData$ECOGL <- factor(exposureData$ECOGL,
levels=c("HV","ECOG: 0/1","ECOG: 2/3"))
# ARAL
exposureData$ARAL[exposureData$ARA %in% c("Yes","Both","Yes*")] <- "Yes"
exposureData$ARAL[exposureData$ARA == "No"] <- "No"
exposureData$ARAL <- factor(exposureData$ARAL, levels=c("No","Yes"))
#------------------- 100 mg BID AUC24hr,ss vs covariates ----------------------
# Cont
png(filename = paste(resCovModelDir,"predAUCCmaxvsAge.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=AGE)) + geom_point(shape=1) +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Age (years)"),
ggplot(exposureData, aes(y=Cmaxss, x=AGE)) + geom_point(shape=1) +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Age (years)"),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsBWT.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=BWT)) + geom_point(shape=1) +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Body weight (kg)"),
ggplot(exposureData, aes(y=Cmaxss, x=BWT)) + geom_point(shape=1) +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Body weight (kg)"),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvseGFR.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=BEGFR)) + geom_point(shape=1)+
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"),
x=expression("eGFR (mL/min/1.73m"^2~")")),
ggplot(exposureData, aes(y=Cmaxss, x=BEGFR)) + geom_point(shape=1)+
labs(y=expression("C"["max,ss"]~"(ng/mL)"),
x=expression("eGFR (mL/min/1.73m"^2~")")),
nrow=1)
dev.off()
# Cat
png(filename = paste(resCovModelDir,"predAUCCmaxvsSex.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=SEXM)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Sex") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=SEXM)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Sex") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
# FemalesAUC <- quantile(exposureData$AUC24hss[exposureData$SEXM=="Female"], p=c(0.25,0.5,0.75))
# MalesAUC <- quantile(exposureData$AUC24hss[exposureData$SEXM=="Male"], p=c(0.25,0.5,0.75))
# FemalesCmax <- quantile(exposureData$Cmaxss[exposureData$SEXM=="Female"], p=c(0.25,0.5,0.75))
# MalesCmax <- quantile(exposureData$Cmaxss[exposureData$SEXM=="Male"], p=c(0.25,0.5,0.75))
png(filename = paste(resCovModelDir,"predAUCCmaxvsRENAL.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=RENAL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Renal impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=RENAL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Renal impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsHEPATIC.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=HEPATIC)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Hepatic impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=HEPATIC)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Hepatic impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsHEPATICL.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=HEPATICL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Hepatic impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=HEPATICL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Hepatic impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsRACE.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=RACE)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Race") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=RACE)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Race") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsRACEL.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=RACEL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Race") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=RACEL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Race") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsEthnicity.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=ETHNIC)) +
geom_boxplot(fill="#a0a0a0",outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Ethnicity") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=ETHNIC)) +
geom_boxplot(fill="#a0a0a0",outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Ethnicity") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsARA.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=ARAL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Acid reducing agent") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=ARAL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Acid reducing agent") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsECOG.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=ECOG)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="ECOG performance status"),
ggplot(exposureData, aes(y=Cmaxss, x=ECOG)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="ECOG performance status"),
nrow=1)
dev.off()
# ggplot(exposureData, aes(y=AUC24hss, x=DOSE)) +
# geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
# stat_summary(fun.data = getNAUC, geom = "label") +
# geom_hline(aes(yintercept = median(exposureData$iAUC100mgBID)), linetype="dashed") +
# labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Dose") +
# theme(axis.text.x = element_text(angle = 45, hjust=1))
AstraZeneca/pmworkbench documentation built on Nov. 18, 2019, 12:51 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
###################################################
# figures.R
#
# Author: Helena Edlund
# Created on: 2017-07-26
# Last modified on:
# Purpose: Produce all project related figures contained in the report
# Dependencies: environmentPrep.R
###################################################
# -------------------------------------------------
# Goodness of fit for base model
# -------------------------------------------------
grid.arrange(
ggplot(baseModelConcData, aes(y=DV, x=PRED)) +
geom_hline(aes(yintercept=0), linetype="dashed", colour="darkgray") +
geom_abline(slope=1, intercept=0) +
geom_point(shape=1, size = 3, alpha=0.4) +
geom_smooth(method="loess") +
coord_cartesian(xlim=c(minConc,maxConc), ylim=c(minConc,maxConc)) +
labs(y="ln(Observed)", x="ln(Population prediction)") +
settings,
ggplot(baseModelConcData, aes(y=DV, x=IPRED)) +
geom_hline(aes(yintercept=0), linetype="dashed", colour="darkgray") +
geom_abline(slope=1, intercept=0) +
geom_point(shape=1, size = 3, alpha=0.4) +
geom_smooth(method="loess") +
coord_cartesian(xlim=c(minConc,maxConc), ylim=c(minConc,maxConc)) +
labs(y="Observed", x="Individual prediction") +
settings,
ggplot(baseModelConcData, aes(y=CWRES, x=TAPD)) +
geom_hline(aes(yintercept=0), color="darkgray") +
geom_point(shape=1, size = 3, alpha=0.4) +
geom_smooth(method="loess") +
coord_cartesian(ylim=c(-maxCWRES,maxCWRES)) +
scale_y_continuous(breaks=seq(from=-20, to=20, by=2)) +
labs(y="CWRES", x="Time after dose") +
settings,
ggplot(baseModelConcData, aes(y=CWRES, x=PRED)) +
geom_point(shape=1, size = 3, alpha=0.4) +
geom_hline(aes(yintercept=0)) +
geom_smooth(method="loess") +
coord_cartesian(ylim=c(-maxCWRES,maxCWRES)) +
scale_y_continuous(breaks=seq(from=-20, to=20, by=2)) +
labs(y="CWRES", x="ln(Population prediction)") +
settings,
nrow=2)
# -------------------------------------------------
# ETA versus covariates for base model
# -------------------------------------------------
baseModelBaseline <- baseModelConcData[!duplicated(baseModelConcData$ID),]
p <-
ggduo(baseModelBaseline,
c('AGE','BWT','BHT','BEGFR','BBILI','BAST'),
c('ETA3',"ETA4"),
types = list(continuous = scatterWithSmooth),
columnLabelsX = c('AGE','BWT','BHT','BEGFR','BBILI','BAST'),
columnLabelsY = c("Random effect of CL/F",'Random effect of Vc/F'))
print(p, leftWidthProportion = 0.35)
p <-
ggduo(baseModelBaseline ,
c('SEXM','RACEL1','ETHNIC','HEALTHGPL','ECOG','DOSE','HEPATICL','ARA'),
c('ETA3',"ETA4"),
types = list(comboVertical = myBoxplot),
columnLabelsX = c('SEX','RACE','ETHNIC','HEALTHGP','ECOG','DOSE','HEPATIC','ARA'),
columnLabelsY = c("Random effect of CL/F",'Random effect of Vc/F'))
print(p, leftWidthProportion = 0.35, bottomHeightProportion = 0.65)
# -------------------------------------------------
# Parameter distributions for base model
# -------------------------------------------------
grid.arrange(
ggplot(baseModelBaseline) +
geom_histogram(aes(x=CL), fill="#737373", colour="#525252", size = 1,bins = 30) +
labs(x='EBE of CL/F', y='Count') + settings,
ggplot(baseModelBaseline) +
geom_histogram(aes(ETA3), fill="#737373", colour="#525252", size = 1,bins = 30) +
# stat_function(fun = dnorm, args = list(sd = sqrt(0.237)), size = 1, linetype = 2) +
labs(x='Random effect of CL/F', y='Count') + settings,
ggplot(baseModelBaseline) +
geom_histogram(aes(x=V2), fill="#737373", colour="#525252", size = 1,bins = 30) +
labs(x='EBE of Vc/F', y='Count') + settings,
ggplot(baseModelBaseline) +
geom_histogram(aes(ETA4), fill="#737373", colour="#525252", size = 1,bins = 30) +
# stat_function(fun = dnorm, args = list(sd = sqrt(0.237)), size = 1, linetype = 2) +
labs(x='Random effect of Vc/F', y='Count') + settings,
nrow=2)
# -------------------------------------------------
# Individual fits for base model
# -------------------------------------------------
## Select 10 best and 10 worst fit based on CWRES.
largestCWRES <-
ddply(baseModelConcData[baseModelConcData$ID !=446,], # ID 446 has only 1 conc
.(ID), summarize,
maxCWRES = max(abs(CWRES), na.rm=T))
largestCWRES <- largestCWRES[order(largestCWRES$maxCWRES),]
best <- head(largestCWRES, n = 12)$ID
worst <- tail(largestCWRES, n = 12)$ID
bestConcData <- baseModelConcData[baseModelConcData$ID %in% best, ]
worstConcData <- baseModelConcData[baseModelConcData$ID %in% worst, ]
# Remove one sparse occasion on a different dosing regimen.
# 2 rich profiles at 200 mg BID ans since plot facets on regimen the sparse
# gets a separate plot area. Does not add much to the figure so excluded.
worstConcData <-
worstConcData[!(worstConcData$ID==337 & worstConcData$Regimen=="100 mg BID"),]
ggplot(data=bestConcData[!is.na(bestConcData$OCC),],
aes(x=TAPD, y=DV, colour=factor(OCC), group=ID.OCC)) +
# observed, rich
geom_point(aes(shape=factor(BLQ)), size=2.5) +
scale_shape_manual(values=c(1,4)) + guides(shape="none") +
# population prediction, rich
geom_line(aes(x=TAPD, y=PRED, group=ID.OCC), colour="#575757",
linetype="solid", inherit.aes = F, size=1) +
# individual prediction, rich
geom_line(aes(x=TAPD, y=IPRED, colour=factor(OCC), group=ID.OCC),
linetype="longdash", size=1) +
# observed, sparse
geom_point(data=bestConcData[is.na(bestConcData$OCC),],
aes(y=DV, x=TAPD), size=2.5, shape=1, col="#575757") +
# predicted, sparse
geom_point(data=bestConcData[is.na(bestConcData$OCC),],
aes(y=IPRED, x=TAPD), size=2.5, shape=2, col="#575757") +
# lloq
geom_hline(aes(yintercept=log(1)), linetype="dashed") +
# settings
coord_cartesian(xlim=c(0,12), ylim=c(0,8)) +
scale_x_continuous(breaks=c(0,2,4,6,8,10,12)) +
scale_colour_brewer(palette="Dark2") +
guides(colour="none") +
facet_wrap(~ID+Regimen, ncol=3, labeller="label_both") +
labs(x="Time after dose (h)", y = "log(Concentration)") + theme_bw() +
settings
# -------------------------------------------------
# Forest plot for the parameter level
# -------------------------------------------------
forrest <-
covParameters[covParameters$par %in% covPar[!covPar %in% c("EGFR_CL","DOSE_CL")],
c("par","estimate","SE")]
forrest$strPar <- gsub("(.*)_","", forrest$par)
forrest$strPar <- factor(forrest$strPar, levels = c("CL", "V"), labels = c("CL/F", "Vc/F"))
forrest$par <- gsub("_.*","", forrest$par)
# estimates and intervals
forrest$estimate <- forrest$estimate+1
forrest$lCI <- forrest$estimate - 1.96*forrest$SE + 1
forrest$uCI <- forrest$estimate + 1.96*forrest$SE + 1
CL <- covParameters$estimate[covParameters$par=="CL"]
V <- covParameters$estimate[covParameters$par=="Vc"]
EGFR <- covParameters$estimate[covParameters$par=="EGFR_CL"]
DOSE <- covParameters$estimate[covParameters$par=="DOSE_CL"]
fEGFR <- data.frame(par="EGFR", estimate=1, SE=NA, strPar="CL/F",
lCI = (CL*(30/75.3)^EGFR)/CL, # min = 29.95
uCI = (CL*(170/75.3)^EGFR)/CL) # max = 280.9 but likely an outlier. next is 168.7
fDOSE <- data.frame(par="Dose", estimate=1, SE=NA, strPar="CL/F",
lCI = (CL*(15/100)^DOSE)/CL,
uCI = (CL*(400/100)^DOSE)/CL)
# allometric scaling
fBWCL <- data.frame(par="BW", estimate=1, SE=NA, strPar="CL/F",
lCI = (CL*(50/70)^0.75)/CL, # min in male patients
uCI = (CL*(140/70)^0.75)/CL) # max in male patients
fBWV <- data.frame(par="BW", estimate=1, SE=NA, strPar="Vc/F",
lCI = (V*(50/70)^1)/V,
uCI = (V*(140/70)^1)/V)
forrest <- rbind(forrest, fEGFR, fDOSE, fBWCL, fBWV)
forrest$par <-
factor(forrest$par,
levels = rev(c("ECOGHV","Dose","EGFR","ARA","SEX","RACEB","RACEO","HEP","BW")),
labels = rev(c("Health group: HV","Dose: 400-15","eGFR: 30-170",
"Acid Reducing Agent",
"Sex: Female",
"Race: Black","Race: Other",
"Hepatic: Mild/Moderate",
"Body Weight: 50-140")))
parImpact <- # printed in section below
ggplot(forrest, aes(estimate, par)) +
geom_rect(xmin = 0.8, ymin = -Inf, xmax= 1.25, ymax = Inf, fill="lightgray") +
geom_point(size=4, shape=18) +
geom_errorbarh(aes(xmax = uCI, xmin = lCI), height = 0.3) +
geom_vline(xintercept = 1, linetype = "longdash") +
theme_bw() + facet_wrap(~strPar) +
scale_x_continuous(breaks = seq(from=0, to=5, by=0.5)) +
labs(x="Change in parameter relative to population estimate", y="") +
theme(axis.title = element_text(size=12),
axis.text = element_text(size=12),
strip.text = element_text(size=10),
legend.title = element_blank(),
legend.text = element_text(size=12),
legend.margin = unit(0.1, "in"),
legend.key.size = unit(0.5, "in"),
legend.position = "bottom",
legend.box = "horizontal",
panel.margin = unit(1.2, "lines"))
# ----------------------------------------------
# Exposure by covariate
# ----------------------------------------------
# function to get numbers in boxes for graphics below
getNAUC <- function(x){
return(c(y = -200, label = length(x)))
}
getNCmax <- function(x){
return(c(y = -40, label = length(x)))
}
theme_set(theme_bw())
res <- 300
## function to divide data into 4 groups base on quartiles
# quartiles <- function(cov) {
# breaks <- quantile(cov, p=c(0.25,0.5,0.75))
# data <- data.frame(cov=cov, grouping=NA)
# data$grouping[data$cov < breaks[1]] <- "0-24.9"
# data$grouping[data$cov >= breaks[1] & data$cov < breaks[2]] <- "25-49.9"
# data$grouping[data$cov >= breaks[2] & data$cov < breaks[3]] <- "50-74.9"
# data$grouping[data$cov >= breaks[3]] <- "75-100"
#
# data$grouping <- factor(data$grouping, levels=c("0-24.9","25-49.9","50-74.9","75-100"))
# return(data$grouping)
# }
### --------- Create missing lumped categories ---------------
# Hepatic
exposureData$HEPATICL[exposureData$HEPATIC == "Normal"] <- "Normal"
exposureData$HEPATICL[exposureData$HEPATIC %in% c("Mild", "Moderate")] <- "Mild/Moderate"
exposureData$HEPATICL <- factor(exposureData$HEPATICL, levels = c("Normal","Mild/Moderate")) # NCI criteria
## Race: white, black, other
exposureData$RACEL[!is.na(exposureData$RACE) & exposureData$RACE == "White"] <- "White"
exposureData$RACEL[!is.na(exposureData$RACE) & exposureData$RACE == "Black/African American"] <- "Black/African American"
exposureData$RACEL[!is.na(exposureData$RACE) & !exposureData$RACE %in% c("White","Black/African American")] <- "Other"
exposureData$RACEL <- factor(exposureData$RACEL, levels=c("White", "Black/African American", "Other"))
# ECOGL
exposureData$ECOGL[exposureData$ECOG %in% c("HV")] <- "HV" # Healty volunteers
exposureData$ECOGL[exposureData$ECOG %in% c("0","1")] <- "ECOG: 0/1" # Healty volunteers
exposureData$ECOGL[exposureData$ECOG %in% c("2","3")] <- "ECOG: 2/3" # Healty volunteers
exposureData$ECOGL <- factor(exposureData$ECOGL,
levels=c("HV","ECOG: 0/1","ECOG: 2/3"))
# ARAL
exposureData$ARAL[exposureData$ARA %in% c("Yes","Both","Yes*")] <- "Yes"
exposureData$ARAL[exposureData$ARA == "No"] <- "No"
exposureData$ARAL <- factor(exposureData$ARAL, levels=c("No","Yes"))
#------------------- 100 mg BID AUC24hr,ss vs covariates ----------------------
# Cont
png(filename = paste(resCovModelDir,"predAUCCmaxvsAge.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=AGE)) + geom_point(shape=1) +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Age (years)"),
ggplot(exposureData, aes(y=Cmaxss, x=AGE)) + geom_point(shape=1) +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Age (years)"),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsBWT.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=BWT)) + geom_point(shape=1) +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Body weight (kg)"),
ggplot(exposureData, aes(y=Cmaxss, x=BWT)) + geom_point(shape=1) +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Body weight (kg)"),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvseGFR.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=BEGFR)) + geom_point(shape=1)+
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"),
x=expression("eGFR (mL/min/1.73m"^2~")")),
ggplot(exposureData, aes(y=Cmaxss, x=BEGFR)) + geom_point(shape=1)+
labs(y=expression("C"["max,ss"]~"(ng/mL)"),
x=expression("eGFR (mL/min/1.73m"^2~")")),
nrow=1)
dev.off()
# Cat
png(filename = paste(resCovModelDir,"predAUCCmaxvsSex.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=SEXM)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Sex") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=SEXM)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Sex") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
# FemalesAUC <- quantile(exposureData$AUC24hss[exposureData$SEXM=="Female"], p=c(0.25,0.5,0.75))
# MalesAUC <- quantile(exposureData$AUC24hss[exposureData$SEXM=="Male"], p=c(0.25,0.5,0.75))
# FemalesCmax <- quantile(exposureData$Cmaxss[exposureData$SEXM=="Female"], p=c(0.25,0.5,0.75))
# MalesCmax <- quantile(exposureData$Cmaxss[exposureData$SEXM=="Male"], p=c(0.25,0.5,0.75))
png(filename = paste(resCovModelDir,"predAUCCmaxvsRENAL.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=RENAL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Renal impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=RENAL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Renal impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsHEPATIC.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=HEPATIC)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Hepatic impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=HEPATIC)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Hepatic impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsHEPATICL.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=HEPATICL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Hepatic impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=HEPATICL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Hepatic impairment") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsRACE.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=RACE)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Race") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=RACE)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Race") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsRACEL.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=RACEL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Race") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=RACEL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Race") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsEthnicity.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=ETHNIC)) +
geom_boxplot(fill="#a0a0a0",outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Ethnicity") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=ETHNIC)) +
geom_boxplot(fill="#a0a0a0",outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Ethnicity") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsARA.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=ARAL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Acid reducing agent") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
ggplot(exposureData, aes(y=Cmaxss, x=ARAL)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="Acid reducing agent") +
theme(axis.text.x = element_text(angle = 45, hjust=1)),
nrow=1)
dev.off()
png(filename = paste(resCovModelDir,"predAUCCmaxvsECOG.png", sep="/"),
width=10, height=4.5, units = "in", res=res)
grid.arrange(
ggplot(exposureData, aes(y=AUC24hss, x=ECOG)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNAUC, geom = "label") +
labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="ECOG performance status"),
ggplot(exposureData, aes(y=Cmaxss, x=ECOG)) +
geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
stat_summary(fun.data = getNCmax, geom = "label") +
labs(y=expression("C"["max,ss"]~"(ng/mL)"), x="ECOG performance status"),
nrow=1)
dev.off()
# ggplot(exposureData, aes(y=AUC24hss, x=DOSE)) +
# geom_boxplot(fill="#a0a0a0", outlier.shape = 1) +
# stat_summary(fun.data = getNAUC, geom = "label") +
# geom_hline(aes(yintercept = median(exposureData$iAUC100mgBID)), linetype="dashed") +
# labs(y=expression("AUC"["24h,ss"]~"(ng/mL*h)"), x="Dose") +
# theme(axis.text.x = element_text(angle = 45, hjust=1))
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.