Changing colour labels of samples with pheatmap

Question

[mar77]

Bit of an R newbie here. I'm trying to generate a figure to see how RNA-seq samples are grouping via hierarchical clustering.

Using this code

rld<-vst(dds, blind=TRUE)
rld_mat<- assay(rld)
rld_cor<-cor(rld_mat)
head(rld_cor)
pheatmap(rld_cor,annotation = meta)
heat.colors<-brewer.pal(9, "Blues")
annotdf<-data.frame(row.names = rownames(rld_cor))
pheatmap(rld_cor, 
         annotation=meta, 
         color=heat.colors,
         annotation_colors = ColorCode,
         border_color = NA, 
         fontsize = 10, 
         fontsize_row = 10, 
         height=20)

But when I try to add this code in:

ColorCode<-c("#CDCDCD","#A5A5A5","#808080","#9AB3FF","#537EFF","#265DFF", "#FFD733","#DDB310", "#CC9900","#FB8ACA","#FB49B0","#E40081","#00FF89","#00B25D","#008042")

and then add this into the pheatmap function

         annotation_colors = ColorCode,

to try and change the colour labels for the samples to match other figures I keep getting this error

Error in annotation_colors[[names(annotation)[i]]] <- factor_colors[ind] : 

more elements supplied than there are to replace

I don't really understand why as I have set up 15 colours and by samples are sorted into 15 categories. Does anyone know how to fix this? Thanks for any help

Answer 1

[Papyrus]

I think the input to annotation_colors should be a list. For example:

ColorCode  = list(sampletype = c("#CDCDCD","#A5A5A5","#808080","#9AB3FF","#537EFF","#265DFF", "#FFD733","#DDB310", "#CC9900","#FB8ACA","#FB49B0","#E40081","#00FF89","#00B25D","#008042"))

And also, I would work with the: annotation_col and annotation_row arguments, not with annotation

Try this example:

library(pheatmap)

foo <- matrix(1:6,2,3)
colnames(foo) <- c("a","b","c")

anno.colors <- list(anno = c(a = "black",b = "green",c = "blue"))

pheatmap(foo,
         annotation_colors  = anno.colors,
         annotation_col = data.frame(anno = c("a","b","c"),
                                     row.names = c("a","b","c")))