Scientific Figure Export

Use this skill when saving figures. These are opinionated defaults to keep files editable and publication-ready.

Downstream use

• •Expect that figures will be edited in Adobe Illustrator, included in latex manuscripts, saved as PDFs, and submitted to scientific journals with design standards.

Save formats

• •Always save SVG + PNG.
• •SVG is the editable master; PNG is the raster preview / export.
• •Set SVG text to editable (matplotlib: svg.fonttype = 'none').

Fonts

• •Use Arial as the default font family.
• •Keep font sizes consistent across panels; avoid tiny text. Most figures will become small panels in a larger figure, and text will be scaled to at least 8pt in the final figure.

Raster quality

• •For PNG exports, use >= 450 DPI (prefer 600 DPI when file size permits). If figure is very large, prompt user for guidance.
• •If a plot is very dense (e.g. scatter plot with many points that would result in >~10MB SVG), prompt user to consider setting rasterized=True for that artist so the SVG stays light but text remains vector.

Transparency & opacity

• •Use alpha/transparency sparingly. It is difficult to increase opacity post-hoc in Illustrator.

Colormaps & normalization

• •Use perceptually uniform colormaps (e.g., viridis) unless there is a strong reason not to. Dichromatic colormaps are great. Use the appropriate class of colormap for the problem.
• •For comparison across panels, use shared normalization (same vmin/vmax). Document if per-panel normalization is used.

Folder structure

• •Save into hierarchical subfolders named after the analysis, e.g.:
  • •figures/consistency_curves/barplot_means
  • •figures/consistency_curves/barplot_medians
  • •figures/consistency_curves/scatter_by_mouse

Recommended helper

• •Look into using the bnpm.plotting_helpers module for figure making and saving. Read relevant docstrings and examples.
• •Use bnpm.plotting_helpers.Figure_Saver:
  • •format_save=['svg', 'png']
  • •svg_fontType='none'
  • •kwargs_savefig={'dpi': 600, 'bbox_inches': 'tight', 'pad_inches': 0.1, 'transparent': False}

Minimal Matplotlib config

import matplotlib as mpl
mpl.rcParams['font.family'] = 'Arial'
mpl.rcParams['svg.fonttype'] = 'none'