Forest plot with HR table

[LK8]

Hello everyone,

I'm still pretty new at R, so I hope this is not an obvious question. I'm currently trying to visualize my data in a forest plot. For each gene, I have 2 HR for survival data, with respective CI and p-value. I would like to present these HR on the same plot in a different color, with a table next to the plot displaying the actual data. This plot I found online displays it well:

I managed to make this through the Forestplot package using its vignette: Introduction to forest plots

# Load package 
library(forestplot)
#> Loading required package: grid
#> Loading required package: magrittr
#> Loading required package: checkmate
# Generate dummy dataset
Genes <- c("Gene1","Gene2", "Gene3", "Gene4", "Gene5")
HR1 <- c(1.0, 1.2, 1.3, 1.4, 1.5)
HR2 <- c(1.2, 1.4, 1.5, 1.6, 1.7)
lower1 <- c(0.9, 1.0, 1.1, 1.2, 1.3)
lower2 <- c(1.0, 1.1, 1.2, 1.3, 1.4)
upper1 <- c(1.4, 1.6, 1.8, 2.0, 2.2)
upper2 <- c(1.7, 1.9, 2.1, 2.3, 2.5)
p.value1 <- c(2e-03, 2e-02, 3e-01, 5e-04, 1e-03)
p.value2 <- c(4e-01, 6e-04, 2e-03, 5e-01, 7e-04)
dummydata <- data.frame(Genes, HR1, HR2, lower1, lower2, upper1, upper2, p.value1, 
                    p.value2)

# Text on plot
tabletextdummy <- cbind(c("Genes",dummydata$Genes))

# Plot
forestplot(tabletextdummy, mean = cbind(c(NA, dummydata$HR1), c(NA,dummydata$HR2)),
                      lower = cbind (c(NA,dummydata$lower1), c(NA,dummydata$lower2)), 
                      upper = cbind(c(NA,dummydata$upper1), c(NA, dummydata$upper2)),
                      new_page = TRUE,
                      clip = c(0.1,5), 
                      lineheight = unit(10,"mm"),
                      line.margin = .1,
                      xlog = TRUE, xlab = "HR with 95% CI", 
                      col = fpColors(box = c("red4", "skyblue3"), 
                                     lines = c("red3", "skyblue2")),
                      fn.ci_norm = c(fpDrawNormalCI, fpDrawDiamondCI),
                      is.summary = c(TRUE,rep(FALSE,5)), 
                      boxsize = 0.4, 
                      xticks = c(0.75, 1, 1.5, 2, 3),
                      legend = c("Group1", "Group2"), 
                      vertices = TRUE)

^{Created on 2021-01-20 by the reprex package (v0.3.0)}

However, when I try to add a table, it renders the 2 HR's per gene next to each other instead of underneath:

# Load package 
library(forestplot)
#> Loading required package: grid
#> Loading required package: magrittr
#> Loading required package: checkmate
# Generate dummy dataset
Genes <- c("Gene1","Gene2", "Gene3", "Gene4", "Gene5")
HR1 <- c(1.0, 1.2, 1.3, 1.4, 1.5)
HR2 <- c(1.2, 1.4, 1.5, 1.6, 1.7)
lower1 <- c(0.9, 1.0, 1.1, 1.2, 1.3)
lower2 <- c(1.0, 1.1, 1.2, 1.3, 1.4)
upper1 <- c(1.4, 1.6, 1.8, 2.0, 2.2)
upper2 <- c(1.7, 1.9, 2.1, 2.3, 2.5)
p.value1 <- c(2e-03, 2e-02, 3e-01, 5e-04, 1e-03)
p.value2 <- c(4e-01, 6e-04, 2e-03, 5e-01, 7e-04)
dummydata <- data.frame(Genes, HR1, HR2, lower1, lower2, upper1, upper2, p.value1, 
                        p.value2)
# Text on plot
tabletextdummy2 <- cbind(c("Genes",dummydata$Genes), 
                         cbind(c("HR", dummydata$HR1),c(NA, dummydata$HR2)), 
                         cbind(c("CI lower", dummydata$lower1), c(NA, dummydata$lower2)),
                         cbind(c("CI higher", dummydata$upper1), c(NA, dummydata$upper2)), 
                         cbind(c("p-value", dummydata$p.value1), c(NA, dummydata$p.value2)))

# Plot
forestplot(tabletextdummy2, 
           mean= cbind(c(NA, dummydata$HR1), c(NA,dummydata$HR2)),
           lower = cbind (c(NA,dummydata$lower1), c(NA,dummydata$lower2)), 
           upper = cbind(c(NA,dummydata$upper1), c(NA, dummydata$upper2)),
           new_page = TRUE,
           clip = c(0.1,5), 
           lineheight = unit(10,"mm"),
           line.margin = .1,
           xlog = TRUE, xlab = "HR with 95% CI", 
           col = fpColors(box = c("red4", "skyblue3"), 
                          lines = c("red3", "skyblue2")),
           fn.ci_norm = c(fpDrawNormalCI, fpDrawDiamondCI),
           is.summary = c(TRUE,rep(FALSE,5)), 
           graph.pos = 2,
           boxsize = 0.4, 
           xticks = c(0.75, 1, 1.5, 2, 3),
           legend = c("Group1", "Group2"), 
           vertices = TRUE)

^{Created on 2021-01-20 by the reprex package (v0.3.0)}

I've tried other strategies, such as combining the forestplot with a table through ggarrange, cowplot etc. These always give an error as forestplot doesn't generate a ggplot format.

Could anyone give me advice on how to solve this, or which other package could accomplish this?

Thank you so much!

[jrkrideau]

I don't know if this would help but there appears to be a ggforestplot. Could you use it instead?

Forestplots of Measures of Effects and Their Confidence Intervals

[MyKo101]

I think your issue is with this part here:

tabletextdummy2 <- cbind(c("Genes",dummydata$Genes), 
                         cbind(c("HR", dummydata$HR1),c(NA, dummydata$HR2)), 
                         cbind(c("CI lower", dummydata$lower1), c(NA, dummydata$lower2)),
                         cbind(c("CI higher", dummydata$upper1), c(NA, dummydata$upper2)), 
                         cbind(c("p-value", dummydata$p.value1), c(NA, dummydata$p.value2)))

The cbind() function takes in vectors are sticks them side-by-side into a matrix. So on each of these lines, you're making a matrix, cbind(c("HR",dummydata$HR1), c(NA,dummydata$HR2)) is a 6x2 matrix. The next line is also a 6x2 matrix.

You've then wrapped all of these matrices in another cbind() with c("Genes", dummydata$Genes) (which is a vector of length 6). So altogether, you've made a 6x9 matrix. Run tabletextdummy2 into the console and you'll see.

For each of the mean, lower and upper arguments to forestplot() you can give a matrix, and forestplot() pieces them together, but in the table, it'll just present it exactly as is.

There doesn't seem to be a way to do this easily in forestplot(), however that doesn't mean impossible, as long as you're only wanting two groups (as in your example).

From the vignette, forestplot() can take expressions and parse them, such as when it displays a beta symbol. We can use the atop() math notation in R to create expressions which look like "a" over "b", like a fraction without the line. So here's a function code to build these expressions:

sfrac <- function(top,bottom,data=NULL)
  with(data,lapply(paste0("atop(",top,",",bottom,")"),str2expression))

This will return a list that contains the relevant expressions. To use it in your code, we just replace the tabletextdummy2 with the following:

tabletextdummy2<- list(
  c("Genes",dummydata$Genes),
  c("HR",sfrac(HR1,HR2,data=dummydata)),
  c("CI lower",sfrac(lower1,lower2,data=dummydata)),
  c("CI upper",sfrac(upper1,upper2,data=dummydata)),
  c("p-value",sfrac(p.value1,p.value2,data=dummydata))
  )

Every line above creates an element for our list(), which will become a column in the end table. Within each of those (apart from the first one) is a list which starts with the relevant title, "HR", "CI lower", etc... and then contains the sfrac() expressions that we've just built.

The only thing you'll then need to change in your forestplot() call is the lineheight argument. I set it to lineheight = unit(20,"mm") (twice yours) and it worked fine.

[LK8]

Thank you so much, that did the job!