anno_mark color labels aren't as expected

[RodrigoGM]

Hi ! I could use some help to figure out what's going awry.

I'm generating an alteration frequency plot with selected amplified or deleted genes annotated up top. I've added a labels_gp(col = gene.colors), and link_gp(col = gene.colors); where gene.colors is a named vector to color code amplified or deleted genes.

However, the colors in the rendered plot are not as expected. Is there a different way to code these ? or any idea of what's happening?

Here's what I'm doing:

## list of genes
## the mark genes function returns the location of each gene in the heatmap matrix
gth <- mark.genes(cnr, gene.list = genes.to.highlight)
## head(gth)
#>   g1    g2    g3    g4    g5    g6
#>  413   874   949  1177  1240  1306

## pull pre-estimated gene frequencies 
gene.fq <- gene.frequencies.all[gsub("-", ".", names(gth)), c("AmpFQ", "delFQ")]

fqa <- gene.frequencies.all$AmpFQ
fqd <- gene.frequencies.all$delFQ

## which gene is amplified or deleted
## which is max (1 = AmpFQ, 2 = delFQ)
wmx <- apply(gene.fq, 1, which.max)

## conditional color coding, if 1 == 'red', 2 == 'blue'
## black otherwise
gene.colors <- ifelse(wmx == 1, "red",
               ifelse(wmx == 2, "blue", "black"))

## check if statments match order of gth
head(gene.colors, n = 11)
#>     g1     g2     g3     g4     g5     g6     g7     g8     g9    g10    g11
#>  "red" "blue" "blue" "blue" "blue" "blue" "blue" "blue"  "red"  "red"  "red"

head(gth, n = 11)
#>   g1    g2    g3    g4    g5    g6    g7    g8    g9   g10   g11   g12   g13
#>  413   874   949  1177  1240  1306  1476  1522  1783  1810  2260  2948  3046

## check all is concordant
head(cbind(gth, gene.fq, wmx, gene.colors), n = 12)
#>         gth      AmpFQ      delFQ wmx gene.colors
#>   g1    413 0.14583333 0.04166667   1         red
#>   g2    874 0.00000000 0.29166667   2        blue
#>   g3    949 0.00000000 0.31250000   2        blue
#>   g4   1177 0.00000000 0.29166667   2        blue
#>   g5   1240 0.00000000 0.29166667   2        blue
#>   g6   1306 0.00000000 0.43750000   2        blue
#>   g7   1476 0.00000000 0.29166667   2        blue
#>   g8   1522 0.00000000 0.56250000   2        blue
#>   g9   1783 0.31250000 0.02083333   1         red
#>   g10  1810 0.31250000 0.02083333   1         red
#>   g11  2260 0.95833333 0.00000000   1         red
#>   g12  2948 0.00000000 0.54166667   2        blue

## generate genome annotation track
## annotated genes up top color coded for Amp / Del 
## Amplification Frequency Plot
## Deletion Frequency Plot
genome_annotation <- HeatmapAnnotation(
    genes =  anno_mark(at = gth, labels = names(gth),
                       link_height = unit(4, "mm"),
                       labels_gp = gpar(fontsize = 5,
                                        col = gene.colors,
                                        fontface = "italic"),
                       link_gp = gpar(col = gene.colors)),
    fqa = anno_barplot(fqa,
                       gp = gpar(col = "#D40000"),
                       axis_param = list(gp=gpar(fontsize = 5)),
                       ylim = c(0, 1)),
    fqd = anno_barplot(fqd,
                       gp = gpar(col = "#318CE7"),
                       axis_param = list(gp = gpar(fontsize = 5),
                                         direction = "reverse"),
                       ylim = c(0, 1)),
    annotation_name_side = "left",
    annotation_name_gp = gpar(fontsize = 6),
    show_legend = FALSE)

## generate heatmpa + alteration summary
out_hm <- Heatmap(cnr,
                  top_annotation = genome_annotation,
                  show_row_names = FALSE,
                  border =  TRUE)

However, the figure once plotted does not show the color codes as expected (Shown above for g1-g12). Here are a few examples where:
We'd expect g1 to be red, g6 and g8 to be blue. which does not match.

Similarly, down the plot, we expect g48 to be red, g49->g59 to be all blue, and g60->67 to be red, and these colors appear inverted.

gene.colors[49:67]
#>    g49    g50    g51    g52    g53    g54    g55    g56    g57    g58    g59
#> "blue" "blue" "blue" "blue" "blue" "blue" "blue" "blue" "blue" "blue" "blue"
#>    g60    g61    g62    g63    g64    g65    g66    g67
#>  "red"  "red"  "red"  "red"  "red"  "red"  "red"  "red"