Is it possible to italicize row names with pheatmap()?

Hello!

Is it possible to italicize the text for row names in pheatmap()?

I tried:

pheatmap(hmdf,
    color= (rainbow(96,start=0.0,end=0.74,alpha=1)),
    border_color = NA,
    show_colnames = TRUE,
    show_rownames = TRUE,
    labels_col=paste0(c("control"," "," ", "653"," "," ", "111"," "," ")),
    angle_col=0,
    labels_row=expression(italic(c("ARP3_ARATH", "HF101_ARATH", "PKL_ARATH", "-",
                                   "UPL1_ARATH", "6GPD3_ARATH", "PNSB3_ARATH", 
                                   "FL3H_MALDO", "-", "FL3H_PETCR", "CALS3_ARATH",
                                   "SRP40_YEAST", "ACC1_ARATH", "RBS2_BRANA", "CNGC5_ARATH",
                                   "MPK9_ARATH", "PHYLO_ARATH", "CO1A1_HUMAN", "FB30_ARATH",
                                   "IFRH_ARATH", "PER45_ARATH", "IRE1A_ARATH", "PLST1_ARATH",
                                   "-", "PMTK_ARATH", "SGO2_ARATH", "P2C14_ARATH", "-",
                                   "PSL4_ARATH", "-", "RBS2_BRANA", "VIP1_MOUSE", "RCA_ARATH",
                                   "CA4_ARATH", "G3PA2_ARATH", "BCA1_ARATH", "CB1C_ARATH",
                                   "TEX10_HUMAN", "CB21_SINAL", "CB5_ARATH", "RVE6_ARATH", 
                                   "-", "-", "YQGF_RHOCB", "IAN9_ARATH", "-", "PRO1_NEUCR",
                                   "-", "ASOL_BRANA", "-", "-", "TIR1_ARATH", "BH106_ARATH",
                                   "EGL1_ARATH", "C3H53_ORYSJ", "PIF1_XENLA", "SBT16_ARATH",
                                   "INO1_SESIN", "APR3_ARATH", "GRDP1_ARATH", "APR3_ARATH",
                                   "APR1_ARATH", "SUT33_ARATH", "APR1_ARATH", "-")))

But I just get a continuous string of italicized row names, even if I add \n after each.

Similarly, I tried making an object, call it rnames, comprised of a string of the above row names. This, likewise, does not work.

Any ideas?

Thank you in advance for your time!

David

r rna-seq

• 11,308 views • link

updated 3.7 years ago by Kind Katydid •

• written 6.9 years ago by david.wheeler •

Awesome! Thank you Kevin!

Can I acknowledge you in the appropriate section when we publish this? You have helped me tremendously with a couple of things. Thanks again!

David

• 0 views • link

written 6.9 years ago by david.wheeler •

Hey David, sure thing, Hope that the reviewers go easy on you, though!

• 0 views • link

written 6.9 years ago by Kevin Blighe

Thanks Kevin, We still have to validate the RNA seq data with qtPCR but I will let you know when we submit... Thanks again!

• 0 views • link

written 6.9 years ago by david.wheeler •

1 answer

Hey, there are different ways to do this. The long-winded way is to simply convert your gene names to their unicode italic equivalents. Here, I randomly change the first gene to italic-bold:

UNICODE

# create random data
data <- replicate(10, rnorm(10))
rownames(data) <- paste0("GENE", c(1:nrow(data)))
colnames(data) <- paste("Sample", c(1:ncol(data)))

# change first gene name
rownames(data)[1] <- "\U1D46E\U1D46C\U1D475\U1D46C \U1D468"

require(pheatmap)
require(RColorBrewer)

pheatmap(data,
  color = colorRampPalette(rev(brewer.pal(n = 7, name = "RdYlBu")))(100),
  show_rownames = TRUE,
  cluster_cols = TRUE,
  cluster_rows = TRUE,
  scale = "row",
  clustering_distance_rows = "euclidean",
  clustering_distance_cols = "euclidean",
  clustering_method = "complete",
  border_color = FALSE,
  cex = 1.5,
  labels_row = rownames(data))

NB - note that I was listening to Simon & Garfunkel (Bridge Over Troubled Water)...

---------------------

expression() and bquote()

The better way is more along the lines of what you were doing already:

# create random data
data <- replicate(10, rnorm(10))
rownames(data) <- paste0("GENE", c(1:nrow(data)))
colnames(data) <- paste("Sample", c(1:ncol(data)))

newnames <- lapply(
  rownames(data),
  function(x) bquote(italic(.(x))))

require(pheatmap)
require(RColorBrewer)

pheatmap(data,
  color = colorRampPalette(rev(brewer.pal(n = 7, name = "RdYlBu")))(100),
  show_rownames = TRUE,
  cluster_cols = TRUE,
  cluster_rows = TRUE,
  scale = "row",
  clustering_distance_rows = "euclidean",
  clustering_distance_cols = "euclidean",
  clustering_method = "complete",
  border_color = FALSE,
  cex = 1.5,
  labels_row = as.expression(newnames))

Kevin

• 0 views • link

written 6.9 years ago by Kevin Blighe

Many thanks for this solution, Kevin!

However, the expression() bquote() method doesn't seem to work for the legend.

newLegend <- lapply(
  c("-2", "-1", "0", "1", "2", "3"),
  function(x) bquote(bold(.(x))))

pheatmap(...
         legend_breaks = c(-2:3),
         legend_labels = as.expression(newLegend))

The columns and rows text are properly turned bold, but the legend text is not bold, it shows the words bold("2"), etc...

Is there any easy way(s) around this?

• 0 views • link

written 3.7 years ago by Kind Katydid •