I prepared the code for forestplot, however I have problem with exporting plot with "≥" / more or equal sign.
library(forestplot)
names6 <- c("Variable",
"A ≥ 4000***",
"B ≥ 50***",
"C**",
"D",
"E***",
"F",
"G*",
"H**",
"I*",
"J***")
coef6 <- c(0.42, 1.58, 1.35, 0.49,
0.46, 0.66, 0.62, 1.34, 0.52, 0.72)
low6 <- c(0.34, 1.29, 1.08, 0.21,
0.33, 0.44, 0.43, 1.08, 0.29, 0.61)
high6 <- c(0.51, 1.93, 1.69, 1.21,
0.64, 0.99, 0.91, 1.66, 0.92, 0.86)
boxsize6 <-c(0.2, 0.2, 0.2,0.2,
0.2,0.2, 0.2,0.2,
0.2, 0.2)
test_data <- data.frame(coef=coef6,
low=low6,
high=high6,
boxsize=boxsize6)
row_names <- cbind(names6,
c("OR",test_data$coef), c("CI-95%", test_data$low), c("CI+95%", test_data$high) )
test_data <- rbind(NA, test_data)
#####FIGURE
forestplot(labeltext = row_names,
mean = test_data$coef, upper = test_data$high,
lower = test_data$low,
is.summary=c(TRUE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE),
boxsize = test_data$boxsize,
zero = 1,
xlog = FALSE,
xlab = "OR (95% CI)",
col = fpColors(lines="black", box="black"),
ci.vertices = TRUE,
xticks = c(0,0.5, 1, 1.5, 2.0),
colgap = unit(0.03,'npc'),
lineheight = unit(1.1,"cm"),
txt_gp=fpTxtGp(label = gpar(cex = 0.8),
title = gpar(cex = 1),
ticks = gpar(cex = 0.6),
xlab = gpar(cex = 0.7)))
However I cannot export plot as a .pdf file with present "≥" / more or equal.
Instead of this, I got sign "=" / equal.
What should I change to get this sign on plot?
Edit:
\u2265 do not work...
This solution might be system-specific. Here is something that works on windows:
library("forestplot")
library("withr")
names6 <- c("Variable",
"A \u2265 4000***",
"B \u2265 50***",
"C**",
"D",
"E***",
"F",
"G*",
"H**",
"I*",
"J***")
### Data & params code here....
with_cairo_pdf('forestplot.pdf',
### Forest plot code here....
forestplot(labeltext = row_names,
....)
)
Related
I wrote a script using the "forestplot" package. I want to group the variables in certain categories, which I would like to show in bold, in order to accentuate those categories. How can i adjust my script, so that only certain rows, i.e Risk factor OR (95% CI), patient characteristics, medication history, comorbidities, surgical history and other are shown in bold? I have two colums and 18 rows. Can someone help me? I would be much grateful!!
My script is as below:
tabletext <- cbind(
c("Risk factor" ,"Patient characteristics","Sex, male*", "Bmi (5 points)",
"Alcohol (5 units)", "Smoking*","Medication history",
"Steroid use", "Anticoagulant use*","Comorbidities",
"COPD GOLD 1/2", "COPD GOLD 3/4", "Other pulmonary disease",
"Surgical history",
"Previous colorectal surgery*",
"Previous abdominal surgery (other)","Other", "HIPEC*"),
c("OR (95% CI)",NA, "1.78 (1.20-2.68)", "1.15 (0.95-1.38)", "1.04 (0.94-1.14)",
"1.78 (1.11-2.80)", NA," 1.40 (0.68-2.67)", "1.55 (1.02-2.32)",NA,
"1.40 (0.70-2.61)", "1.56 (0.42-4.67)", "1.78 (0.63-4.28)",NA,
"1.61 (1.03-2.49)", "0.80 (0.47-1.32)",NA, "4.14 (2.14-7.73)"))
?fpTxtGp
require(forestplot)
forestplot(tabletext,
txt_gp = fpTxtGp(label = list(gpar(fontfamily = "Times",
fontface="bold"),
gpar(fontfamily = "",
col = "black"))),
df_c,new_page = TRUE,
boxsize = 0.2,
is.summary = c(rep(FALSE,32)),
clip = c(0,17),
xlab = 'Odds ratio with 95% confidence interval
* indicates significance',
xlog = FALSE,
zero = 1,
plotwidth=unit(12, "cm"),
colgap=unit(2, "mm"),
col = fpColors(box = "royalblue",
line = "darkblue",
summary = "royalblue"))
Its not clear what df_c is so I just created it based on your tabletext matrix:
df_c <- data.frame(mean = c(NA, NA, 1.78, 1.15, 1.04, 1.78, NA, 1.4, 1.55,
NA, 1.4, 1.56, 1.78, NA, 1.61, 0.8, NA, 4.14),
lower = c(NA, NA, 1.2, 0.95, 0.94, 1.11, NA, 0.68, 1.02, NA, 0.7,
0.42, 0.63, NA, 1.03, 0.47, NA, 2.14),
upper = c(NA, NA, 2.68, 1.38,1.14, 2.8, NA, 2.67,2.32, NA,
2.61, 4.67, 4.28, NA, 2.49, 1.32, NA, 7.73))
From there, its just a matter of adjusting the values passed to is.summary:
forestplot(tabletext,
txt_gp = fpTxtGp(label = list(gpar(fontfamily = "Times"),
gpar(fontfamily = "",
col = "black"))),
df_c,new_page = TRUE,
boxsize = 0.2,
is.summary = c(TRUE, TRUE, rep(FALSE, 4),
TRUE, FALSE, FALSE, TRUE,
rep(FALSE,3), TRUE, rep(FALSE,4)),
clip = c(0,17),
xlab = 'Odds ratio with 95% confidence interval
* indicates significance',
xlog = FALSE,
zero = 1,
plotwidth=unit(12, "cm"),
colgap=unit(2, "mm"),
col = fpColors(box = "royalblue",
line = "darkblue",
summary = "royalblue"))
Which generates the following figure:
I would like to clip (or maybe the right formulation is performing spatial intersection) polygons and lines using a polygon rather than a rectangle, like so:
Here is some code to make the polygons for reproducibility and examples:
p1 <- data.frame(x = c(-0.81, -0.45, -0.04, 0.32, 0.47, 0.86, 0.08, -0.46, -1, -0.76),
y = c(0.46, 1, 0.64, 0.99, -0.04, -0.14, -0.84, -0.24, -0.44, 0.12))
p2 <- data.frame(x = c(-0.63, -0.45, -0.2, -0.38, -0.26, -0.82, -0.57, -0.76),
y = c(-0.1, 0.15, -0.17, -0.79, -1, -0.97, -0.7, -0.61))
l1 <- data.frame(x = c(0.1, 0.28, 0.29, 0.52, 0.51, 0.9, 1),
y = c(0.19, -0.15, 0.25, 0.28, 0.64, 0.9, 0.47))
plot.new()
plot.window(xlim = c(-1, 1), ylim = c(-1,1))
polygon(p2$x, p2$y, col = "blue")
polygon(p1$x, p1$y)
lines(l1$x, l1$y)
You could use the spatstat package for this. Below the original example is
worked through. In spatstat polygons are used as “observation windows” of
point patterns, so they are of class owin. It is possible to do set
intersection, union etc. with owin objects.
p1 <- data.frame(x = c(-0.81, -0.45, -0.04, 0.32, 0.47, 0.86, 0.08, -0.46, -1, -0.76),
y = c(0.46, 1, 0.64, 0.99, -0.04, -0.14, -0.84, -0.24, -0.44, 0.12))
p2 <- data.frame(x = c(-0.63, -0.45, -0.2, -0.38, -0.26, -0.82, -0.57, -0.76),
y = c(-0.1, 0.15, -0.17, -0.79, -1, -0.97, -0.7, -0.61))
l1 <- data.frame(x = c(0.1, 0.28, 0.29, 0.52, 0.51, 0.9, 1),
y = c(0.19, -0.15, 0.25, 0.28, 0.64, 0.9, 0.47))
In spatstat polygons must be traversed anti-clockwise, so:
library(spatstat)
p1rev <- lapply(p1, rev)
p2rev <- lapply(p2, rev)
W1 <- owin(poly = p1rev)
W2 <- owin(poly = p2rev)
L1 <- psp(x0 = l1$x[-nrow(l1)], y0 = l1$y[-nrow(l1)],
x1 = l1$x[-1], y1 = l1$y[-1], window = boundingbox(l1))
plot(boundingbox(W1,W2,L1), type= "n", main = "Original")
plot(W2, col = "blue", add = TRUE)
plot(W1, add = TRUE)
plot(L1, add = TRUE)
W2clip <- W2[W1]
L1clip <- L1[W1]
plot(W1, main = "Clipped")
plot(W2clip, col = "blue", add = TRUE)
plot(L1clip, add = TRUE)
I'd be grateful for your help. I'm using the forestplot R package to create a forest plot. It was going well.
library(forestplot)
test_data <- data.frame(
coef = c(1.53, 1.56, 1.11, 1.35, 1.00, 0.94, 1.03),
low = c(1.43, 1.38, 1.08, 1.05, 0.91, 0.87, 0.95),
high = c(1.64, 1.76, 1.15, 1.75, 1.10, 1.01, 1.11),
boxsize = c(0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1)
)
row_names <- cbind(
c("Variable", "BMI", "WHR", "Diabetes", "Insulin resistance", "LDL", "HDL", "ApoB"),
c("SNPs", "807", "368", "386", "53", "213", "510", "241"),
c("OR (95% CI)", "1·53 (1·43−1·64)", "1·56 (1·38−1·76)", "1·11 (1·08−1·15)", "1·35 (1·05−1·75)", "1·00 (0·91−1·10)", "0·94 (0·87−1·01)", "1·03 (0·95−1·11)"),
c("p-value", "6·72E−35", "2·90E−12", "6·61E−10", "0·02", "0·97", "0·083", "0·53")
)
test_data <- rbind(rep(NA, 3), test_data)
forestplot(
labeltext = row_names,
xlab = "Odds ratio",
test_data[, c("coef", "low", "high")],
is.summary = FALSE,
boxsize = test_data$boxsize,
zero = 1,
cex = 0.1,
align = TRUE,
xlog = FALSE,
graphwidth = unit(1.5, "snpc"),
col = fpColors(lines = "royalblue", box = "darkblue")
)
I've been adding one variable at a time and plotting my graph after the addition of each to make sure there were no errors introduced at each step. It was all going well until I added my eighth variable, "Triglycerides", and its associated numbers:
library(forestplot)
test_data <- data.frame(
coef = c(1.53, 1.56, 1.11, 1.35, 1.00, 0.94, 1.03, 1.15),
low = c(1.43, 1.38, 1.08, 1.05, 0.91, 0.87, 0.95, 1.07),
high = c(1.64, 1.76, 1.15, 1.75, 1.10, 1.01, 1.11, 1.11),
boxsize = c(0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1)
)
row_names <- cbind(
c("Variable", "BMI", "WHR", "Diabetes", "Insulin resistance", "LDL", "HDL", "ApoB", "Triglycerides"),
c("SNPs", "807", "368", "386", "53", "213", "510", "241", "427"),
c("OR (95% CI)", "1·53 (1·43−1·64)", "1·56 (1·38−1·76)", "1·11 (1·08−1·15)", "1·35 (1·05−1·75)", "1·00 (0·91−1·10)", "0·94 (0·87−1·01)", "1·03 (0·95−1·11)", "1·15 (1·07−1·11)"),
c("p-value", "6·72E−35", "2·90E−12", "6·61E−10", "0·02", "0·97", "0·083", "0·53", "2·14E−4")
)
test_data <- rbind(rep(NA, 3), test_data)
forestplot(
labeltext = row_names,
xlab = "Odds ratio",
test_data[, c("coef", "low", "high")],
is.summary = FALSE,
boxsize = test_data$boxsize,
zero = 1,
cex = 0.1,
align = TRUE,
xlog = FALSE,
graphwidth = unit(1.5, "snpc"),
col = fpColors(lines = "royalblue", box = "darkblue")
)
On adding this eighth variable, I now receive the following error message:
Error in prFpConvertMultidimArray(mean) :
Sorry did not manage to correctly identify the upper/lower boundaries from the input matrix.
I can't spot a syntax error here, and I'm really at a loss as to why this has only just happened. Any advice would be greatly appreciated.
Thank you.
Please, I need some help about using the xlog=TRUE option.
It is requested to provide the mean, lower, upper, zero, grid and clip already as exponentials, but I find the package is drawing the grid lines at the exponential of the numbers I am already providing as exponentials. As a consequence, the grid lines are in the wrong place.
metaan <-
structure(list(
mean = c(NA, NA, NA, 0.27, 0.47, 0.33, 0.69, 0.86, 0.37, 0.08, 0.44, 0.54, 0.41, NA),
lower = c(NA, NA, NA, 0.13, 0.12, 0.19, 0.12, 0.54, 0.17, 0.03, 0.16, 0.06, 0.29, NA),
upper = c(NA, NA, NA, 0.58, 1.81, 0.60, 3.97, 1.36, 0.81, 0.21, 1.25, 4.50, 0.58, NA)),
.Names = c("mean", "lower", "upper"),
row.names = c(NA, -27L),
class = "data.frame")
tabletext<-cbind(
c("", "AB class", "", " Aminoglycosides", " B-lactams", " Cephalosporins", " Fenicoles", " Fluoroquinolones", " Multiresistance", " Sulphamides", " Tetracyclines", " Tri/Sulpha", " Subtotal", ""),
c("", "OR", "", "0.27", "0.47", "0.33", "0.69", "0.86", "0.37", "0.08", "0.44", "0.54", "0.41", ""),
c("", "n", "", "4", "3", "2", "3", "4", "2", "3", "4", "3", "5", ""))
xticks <- c(0.1, 0.25, 0.5, 1, 1.5, 2, 3)
forestplot(tabletext,
graph.pos = 3,
txt_gp = fpTxtGp(label = gpar(fontsize=10)),
hrzl_lines = list("3" = gpar(lty=1)),
zero = 1,
line.margin = .05,
mean = cbind(metaan[,"mean"]),
lower = cbind(metaan[,"lower"]),
upper = cbind(metaan[,"upper"]),
is.summary=c(FALSE, TRUE, rep(FALSE, 9)),
col=fpColors(box=c("blue"), summary=c("blue")),
grid = structure(0.41,
gp = gpar(lty = 2, col = "#CCCCFF")),
clip=c(0.1, 3),
xlog=T,
xticks=xticks,
xlab="Odds ratio")
The grid line is at the exponential of OR=0.41, instead of at OR=0.41
When provided the log to get the grid lines at the correct place (e.g. -0.38, or log(0.41)), I get the error message that I should provide all parameters already as exponential.
forestplot(tabletext,
graph.pos = 3,
txt_gp = fpTxtGp(label = gpar(fontsize=10)),
hrzl_lines = list("3" = gpar(lty=1)),
zero = 1,
line.margin = .05,
mean = cbind(metaan[,"mean"]),
lower = cbind(metaan[,"lower"]),
upper = cbind(metaan[,"upper"]),
is.summary=c(FALSE, TRUE, rep(FALSE, 9)),
col=fpColors(box=c("blue"), summary=c("blue")),
grid = structure(-0.39,
gp = gpar(lty = 2, col = "#CCCCFF")),
clip=c(0.1, 3),
xlog=T,
xticks=xticks,
xlab="Odds ratio")
Error in forestplot.default(tabletext, graph.pos = 3, txt_gp = fpTxtGp(label = gpar(fontsize = 10)), :
All argument values (mean, lower, upper, zero, grid and clip) should be provided as exponentials when using the log scale. This is an intentional break with the original forestplot function in order to simplify other arguments such as ticks, clips, and more.
I have tried including the grid numbers as lists, but I always encounter the same error message either if I provide the numbers as exponentials (grid misplaced) or as log (error message).
I am wondering what I am doing wrong and if there is any other way to get the grid lines in the correct place.
Thanks in advance,
Magda.
Solved. Updated to 1.8 package version from GitHub and got the correct figure. :)
I have made a forest plot in r using the forestplot package. The code is as follows:
#forestplot
labeltext2 <- c("A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N", "O", "P", "Q", "R", "S", "Summary Effect")
effect2 <- c(0.12, 0.61, 0.11, 0.25, 0.24, 0.63, 0.33, 0.41, 0.38, 0.52, 0.23, 0.47,
0.32, 0.36, 0.03, 0.15, 0.25, 0.67, 0.003, 0.32)
lower_2 <- c(0.08, 0.51, .03, 0.11, 0.06, 0.62, 0.11, 0.29, 0.18,
0.4, 0.19, 0.39, 0.24, 0.39, -0.13, 0.01, 0.09, 0.43,
-0.08, 0.19)
higher_2 <- c(0.16, 0.71, .19, 0.39, 0.42, 0.64, 0.55, 0.53, 0.58,
0.64, 0.27, 0.55, 0.4, 0.4, 0.19, 0.29, 0.41, 0.91,
0.08, 0.44)
forestplot(labeltext2, effect2, lower_2, higher_2, zero = .32,
cex = 2,
lineheight = "auto",
xlab = "effect size",
xticks = c(-.5, 0, .5, 1, 1.5),
title = "ForestPlot",
new_page = TRUE)
Which allows me to get this image:
This image has an effect line coming up at .32, the summary effect, using the zero argument. I would like to add an additional thick black line at 0 to show the "no-effect" line. Does anyone know how do this? I am open to using another package/function.
Thank you!
I can't figure out how the forestplot uses grid graphics, but here's a manual solution:
forestplot(labeltext2, effect2, lower_2, higher_2, zero = .32,
cex = 2,
lineheight = "auto",
xlab = "effect size",
xticks = c(-.5, 0, .5, 1, 1.5),
title = "ForestPlot",
new_page = TRUE)
# Add line manually
x = .397
y0 = .12
y1 = .875
grid.lines(c(x, x), c(y0, y1), default.units = "npc",
gp = gpar(lwd = 2))