R-object for the figure exported directly as numerical datatable.

How the pie chart was generated: The pie chart in the manuscript was created in R using ggplot2 by drawing a stacked bar (geom_col()) and converting it to polar coordinates (coord_polar(theta = "y")). The attached CSV was exported from ggplot_build(p)$data[[1]] and therefore contains ggplot2’s internal computed coordinates for the plotted geometry.

Each row corresponds to one slice of the pie chart (one stacked segment before conversion to polar coordinates).

Key columns:
• ymin and ymax: lower and upper cumulative boundaries (in the original stacked bar) for that slice.
These are the primary values that define the slice size.
• Slice size used in the pie chart:
\text{slice\_proportion} = ymax - ymin
The slice angle in the pie is proportional to slice_proportion.
• y: ggplot2’s internal y value for the segment. In this exported table, y matches the cumulative position used during stacking and is not an independent measurement. The relevant slice size is ymax - ymin.

Other columns:
• fill: the computed hexadecimal fill color (after applying scale_fill_gradientn(...)). This corresponds to the visual color of each slice in the manuscript.
• x, xmin, xmax: x position and bar width parameters. In a pie chart created from a single stacked bar, x is constant (here x = 1) and xmin/xmax set the bar thickness.
• PANEL: facet panel index. This figure has one panel (PANEL = 1).
• group: grouping identifier used internally by ggplot2 (constant here).
• linewidth, linetype, alpha, colour, flipped_aes: internal rendering parameters used by ggplot2.

Because this CSV contains ggplot2 internal geometry, it does not include the original biological/statistical variables used to construct the plot (for example category labels such as stat, the original numeric values such as proportion, or the value mapped to color such as log2 odds ratio). The fill column records the final computed color after the gradient mapping.