Italicise annotations with pheatmap
1
0
Entering edit mode
16 months ago
jamie.pike ▴ 80

Hello,

I have some column annotations in a heatmap (generated with pheatmap), which correspond to species that i want to be in italics.

I have read through the pheatmap documentation and I can't find anything. I have also tried to find and adapt solutions online (for example), but most tutorials and posts refer to italicising the row or column names, not the annotations in the legend. I have tried to apply some of these techniques to my script but with no luck.

Is anyone able to suggest a solution? I have seen lots of posts about complex heatmaps, but from what I can work out it is not so easy to do with that either. Surely having species or genes in the annotations, which should technically be italicised, is common?


Here is what I have so far:

#Create a list containing the colours we want to use for the annotations. 
ann_colours = list(
  Race= c("Race 1" = "orange2", "Race 4" = "blue"),
  "Forma_specialis" = c("apii" = "lightskyblue", "cepae" = "purple", "conglutinans" = "orange", "coriandrii" = "pink", "cubense" = "khaki1", "lactucae" = "green3", "lini" = "indianred", "lycopersici" = "red", "matthiolae" = "blue", "narcissus" = "yellow", "niveum" = "brown", "rapae" = "dark green", "rocket" = "light green", "vasinfectum" = "greenyellow", "non-pathogenic" = "black"), 

newnames_rows <- lapply(
  colnames(HeatmapData_Binary),
  function(x) bquote(italic(.(x)))) #This works perfectly for the rows of course - but it's not what I need!

#Try to apply to the annotations
newnames <- lapply(
  list(ann_colours),
  function(x) bquote(italic(.(x)))) 

If I print newnames:

italic(list(Race = c(`Race 1` = "orange2", `Race 4` = "blue"), 
    Forma_specialis = c(apii = "lightskyblue", cepae = "purple", 
    conglutinans = "orange", coriandrii = "pink", cubense = "khaki1", 
    lactucae = "green3", lini = "indianred", lycopersici = "red", 
    matthiolae = "blue", narcissus = "yellow", niveum = "brown", 
    rapae = "dark green", rocket = "light green", vasinfectum = "greenyellow", 
    `non-pathogenic` = "black"), Lactucae_Distribution = c(`Race 1 Specific` = "tan1", 
    `Race 4 Specific` = "skyblue3", Shared = "green", `Not Race Specific` = "grey"
    ))) 

I imagine the issue is because I am not trying to italicise the information in the df, but instead the 'ann_colours' list?

Here is my pheatmap script:

HeatmapData_Binary %>%
  pheatmap(color = colorRampPalette(c("grey90", "firebrick"))(2),
           legend = T,
           legend_breaks = c(0.75, 0.25),
           legend_labels = c("Present", "Absent"),
           drop_levels = F,
           cluster_rows = CEC_Clustering,
           cluster_cols = Fsp_Clustering,
           cellheight = 5.5,
           cellwidth = 60,
           annotation_row = RowAnnotations,
           annotation_col = ColAnnotations,
           annotation_colors = newnames,
           fontsize_row = 2.5,
           fontsize = 24,
           treeheight_row = 10,
           angle_col = 90,
           cutree_cols = 14,
           show_rownames = F,
           na_col = "white",
           border_color = "grey90",
           labels_col = Isolist,
           filename = "./FolaEffectorsHeatmap-Draft1.pdf",
           height = 30,
           width = 25)

Pheatmap which does not contain italicised annotations:

image: pheatmap

I am so close to having this figure finished! This has been plaguing me for days! Please help!

Thank you, Jamie!

R Pheatmap • 1.6k views
ADD COMMENT
2
Entering edit mode

I might be the party pooper but I generally think that these sorts of aesthetics should be done later in something like Inkscape. In R do the figure, in appropriate dimensions and font size and then do everything to make it look publication-ready later. No need to waste time trying to code up everything in R. You're producing dozens or hundreds of plots during a project, and only a few percent will make the paper, not worth losing too much time doing the beauty work on plots that will never again see any daylight.

ADD REPLY
0
Entering edit mode

@ATpoint Very true! I'm not so familiar with many image editing software and I wonder if it will be just as quick for me to try in R, but I think for some of the stuff I want to achieve with this figure that will be the best bet! Thank you (:

ADD REPLY
1
Entering edit mode

Inkscape you learn in a day, for basic image and figure manipulation it's really simple and a good skill to have. Highly recommended.

ADD REPLY
4
Entering edit mode
16 months ago

You might not want to hear this...but you can probably do it with ComplexHeatMap. I put superscripts in ComplexHeatmap, so you should be able to do italics.

ADD COMMENT
1
Entering edit mode

Just want to add that ComplexHeatMap > pheatmap for most things and it's very easy to port code over.

From version 2.5.2 of ComplexHeatmap, I implemented a new ComplexHeatmap::pheatmap() function which actually maps all the parameters in pheatmap::pheatmap() to proper parameters in ComplexHeatmap::Heatmap(), which means, it converts a pheatmap to a complex heatmap.

ADD REPLY
0
Entering edit mode

You're right, I was trying to save time and not write for in ComplexHeatmap, but using @Trivas ComplexHeatmap::pheatmap() should help! Thank you both (:

ADD REPLY

Login before adding your answer.

Traffic: 1904 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6