Ersilia Book
Search…
⌃K

Scientific figures with Stylia

Sytlia is a small Python library for styling plots
​Stylia is a small package to stylize Matplotlib plots in Python so that they are publication-ready. Stylia provides modified axes (ax) that can be used as drop-in replacements for Matplotlib axes.

Getting started

Installation

First make sure that you have the Arial font installed in your computer (Linux systems do not have it preinstalled). The best is to install Arial in the conda environment you are using:
conda install -c conda-forge mscorefonts
You can read more about fonts and Matplotlib in this excellent blogpost from the Alexander Lab.
Stylia is constantly evolving, so we recommend that you install it directly from the GitHub repository.
git clone https://github.com/ersilia-os/stylia.git
cd stylia
pip install -e .

Create a single panel figure

import stylia as st
import numpy as np
​
fig, axs = st.create_figure(1,1)
ax = axs[0]
​
x = np.random.normal(size=100)
y = np.random.normal(size=100)
​
ax.scatter(x, y)
​
st.save_figure("my_single_plot.png")

Create a multipanel figure

import stylia as st
import numpy as np
​
# create a figure to be used in a slide
fig, axs = st.create_figure(nrows=2, ncols=2, width_ratios=[2, 1])
​
# get data
x = np.random.normal(size=100)
y = np.random.normal(size=100)
​
# first plot, access with flat subplots coordinates (0)
ax = axs[0]
# a default color is used
ax.scatter(x, y)
# write labels to axis, title and numbering of the subplot
st.label(
ax,
title="My first plot",
xlabel="This is the X axis",
ylabel="This is the Y axis",
abc="A",
)
​
# second plot, acces with subplots 2D coordinates (0,1)
ax = axs[0, 1]
# use a named color
named_colors = st.NamedColors()
​
def my_scatterplot(ax, x, y):
ax.scatter(x, y, color=named_colors.get("red"))
​
my_scatterplot(ax, x, y)
# write only a new title (the rest are defaults)
st.label(ax, title="My second plot")
​
# third plot, access with next() method
ax = axs.next()
cmap = st.ContinuousColorMap()
cmap.fit(x)
colors = cmap.get(x)
ax.scatter(x, y, color=colors)
​
# fourth plot
ax = axs[1, 1]
# add transparency
ax.scatter(x, y, color=named_colors.get("blue", alpha=0.2))
​
# save figure
st.save_figure("my_grid_plot.png")

Sizes

Figure size

We follow the Nature Figure Guidelines. Please read those style guidelines carefully. In brief, the entire figure should be have the following sizes:
  • SINGLE_COLUMN_WIDTH: 90 mm or 3.54 in
  • TWO_COLOUMNS_WIDTH: 180 mm or 7.09 in
These variables are built-in within Stylia. You can access them as follows:
from stylia import TWO_COLUMNS_WIDTH

Font size

  • FONTSIZE_SMALL: 5
  • FONTSIZE: 6
  • FONTSIZE_BIG : 8

Marker sizes

  • MARKERSIZE_SMALL: 5
  • MARKERSIZE: 10
  • MARKERSIZE_BIG: 30

Line widths

  • LINEWIDTH: 0.5
  • LINEWIDTH_THICK: 1.0

Colors

Named colors

You can use predefined colors:
from stylia import NamedColors
​
named_colors = NamedColors()
color = named_colors.get('blue')
Available color names are:
  • 'red'
  • 'blue'
  • 'green'
  • 'orange'
  • 'purple'
  • 'yellow'
  • 'gray'
  • 'white'
  • 'black'

Color maps

Continuous color maps

Color maps can be created with the fit method.
from stylia import ContinuousColorMap
import numpy as np
​
cmap = ContinuousColorMap("spectral")
x = np.random.normal(size=100)
y = np.random.normal(size=200) / 2
cmap.fit(x)
# get colors of x
colors_x = cmap.get(x)
# get colors of y based on the x scale
colors_y = cmap.get(y)
Available color maps are:
  • 'spectral'
  • 'viridis'
  • 'coolwarm'

Discrete colormaps

Discrete colormaps are work in progress
Please note that, by default, we use Scientific Color Maps. Non-scientific color maps look brighter, though. If you want to use non-scientific color maps, simply specify ContinuousColorMap("spectral", scientific=False).