How to get a legend from genoplotR?

Question

[marongiu.luigi]

I have a genome that I would like to compare to its reference. I have used genoplotR as implemented within fastANI and I got a decent graph:

Is it possible to add a scale to give a sense of the colors? In the manual, scale=TRUE by default and legend is still not implemented...

Thank you

[antonioggsousa]

Hi,

I'm not familiar with this package or function, but from the function documentation that you had provided in the link, it says about the parameter legend: Yet unimplemented.

Answer 1

[Kevin Blighe]

The legend parameter in the plot_gene_map function from genoplotR is not implemented, as stated in the documentation. The scale parameter refers to a genomic position scale, not a color scale. Since genoplotR uses the grid graphics system, you can manually add a color legend by creating an additional viewport and drawing a gradient bar or discrete color squares using grid functions.

If your plot uses a global color scheme based on a comparison metric such as sequence identity, you can add a continuous color legend with code similar to the following. Adjust the colors and values to match your global_color_scheme argument.

library(grid)
library(genoPlotR)

# Your original plot call, with plot_new = FALSE to avoid new page
pushViewport(viewport(layout = grid.layout(1, 2, widths = unit(c(0.8, 0.2), "npc"))))
pushViewport(viewport(layout.pos.col = 1))
plot_gene_map(dna_segs = your_dna_segs, comparisons = your_comparisons, plot_new = FALSE, ...)  # Your parameters here
upViewport()

# Add legend in second column
pushViewport(viewport(layout.pos.col = 2))
# Example for a blue-red gradient legend assuming values from 0 to 100
col_fun <- colorRampPalette(c("blue", "red"))
n <- 100
y <- seq(0, 1, length = n)
grid.rect(x = 0.5, y = y, width = 0.2, height = 1/n, gp = gpar(fill = col_fun(n), col = NA), just = "center")
grid.text(c("0", "50", "100"), x = 0.8, y = c(0, 0.5, 1), just = "left", gp = gpar(cex = 0.8))
grid.text("Identity (%)", x = 0.5, y = 1.1, just = "center", gp = gpar(cex = 0.8))
upViewport(2)

This creates a side legend. For discrete colors, replace the gradient with individual rectangles. Test this in your FastANI visualization script by modifying visualize.R accordingly.

Kevin