R: Extra-space in label position. gplots - VennDiagram - venn.diagramm()

Question

I'm facing this little layout problem here with this 3-vector list venn diagram. I specified cat.pos = c(0, 0, 0), cat.dist = c(0.1, 0.1, 0.1) to have all 3 labels at noon above their respective area and at the same distance but I got the 3rd 1 step more above.. It is also farther when no cat.dist vector is passed to the function.

# Get gene lists to do Venn diagram
ids_3883_pvsUNR <- c(paste0("gene",c(1:66)))
ids_3883_pvsCRSP <- c(paste0("gene",c(67:126)))
ids_3884 <- c(paste0("gene",c(1:30,67:85,1001:4054)))

head(ids_3883_pvsUNR)

length(ids_3883_pvsUNR)
length(ids_3883_pvsCRSP)
length(ids_3884)


# 
geneLS <- list(ids_3883_pvsUNR, ids_3883_pvsCRSP, ids_3884)
# We can rename our list vectors
names(geneLS) <- c("Set_1_3883_pvsUNR","Set_1_3883_pvsCRSP", "Set_2_3884RUV")

# Now we can plot a Venn diagram with the VennDiagram R package, as follows:
require("VennDiagram")

VENN.LIST <- geneLS
labels <- c("Set_1_3883\nPat 1,2 vs UNR","Set_1_3883\nPat 1,2 vs 3 CRSP", "Set_2_3884RUV\nPat 3 vs 3 CRSP")
venn.plot <- venn.diagram(VENN.LIST , NULL, fill=c("darkmagenta", "darkblue", "grey"), alpha=c(0.5,0.5,0.5), cex = 2, cat.fontface=4, cat.pos = c(0, 0, 0), cat.dist = c(0.1, 0.1, 0.1), category.names=labels, main="Set_1_May_3883 (P vs UNR or CRSP) vs Set_2_May_3884_RUVcorrected")

# To plot the venn diagram we will use the grid.draw() function to plot the venn diagram
dev.off()
grid.draw(venn.plot)

Does somebody know what is happening here? Thanks for your feedback! Daniel

Answer 1

Try the functions of the ggVennDiagram or ggvenn package. Here are the data

genes <- paste("gene",1:3180,sep="")
x <- list(
  `Set_1_3883\n Pat 1,2 vs UNR` = genes[1:66], 
  `Set_1_3883\n Pat 1,2 vs 3 CRSP` = genes[37:3139], 
  `Set_2_3884RUV\n Pat 3 vs 3 CRSP` = genes[3121:3180]
)

And now the graph

library(ggVennDiagram)
ggVennDiagram(x, label_alpha = 0)+
  scale_fill_gradient(low="gray",high = "blue")

And with the ggvenn package

library(ggvenn)
ggvenn(
  x, 
  fill_color = c("plum1", "blueviolet", "gray"),
  stroke_size = 0.5, set_name_size = 4
)

In both cases they are ggplots so you will have much better control over them.