A custom R pipeline that generates spatial heatmaps of wing damage from insect wing image data to visualize spatial location of wing damage.
Motivation
While investigating the effect of wing damage on butterfly flight performance during my PhD, I aimed at quantifying not only the extent of damage (e.g., percentage of missing wing area), but also its spatial distribution across the wing surface.
How It Works
I superimposed standardized images of the wings from wild specimens using EBImage R package. This allowed counting the missing area at the pixel scale. The heatmap is then built from a matrix summing the occurrences of missing pixels, and plotted with autoimage R package.
The pixel matrix shown here is at very low resolution for the sake of simplicity. Note that the natural shape variation between individuals (i.e. not due to wing damage) is eliminated in the process to match the intact template.
Code
How It’s Been Used
Check out the use of Wing Damage Heatmap Generator in my research article published in the Journal of Experimental Biology .
Another interesting example is this study on the evolution of butterfly wing tails by Ariane Chotard, published in the Proceedings of the Royal Society B .
