pyanatomogram 0.1.1

PyAnatomogram is a python package to annotate and highlight organs and tissues in figures

Table of contents

[TOC]

What is PyAnatomogram?

PyAnatomogram is a python package to annotate and highlight organs and tissues in figures.

PyAnatomogram parses anatomy svg images from the EBI Gene Expression Group's anatomogram repository and provides a python interface for programmatically annotating the individual organs with colors or other style attributes. See ebi-gene-expression-group anatomogram repository for a for a full list of animal and plant organisms that are supported.

This package was inspired by the gganatogram R-package by Jesper Maag, and was written to provide similar functionality in python and to provide full resolution figures.

Example use - Gallus Gallus (chicken)

Highlight individual organs

Let's color-code the heart and lung in a chicken image according to some measurement of interest.

import pyanatomogram
anatomogram = pyanatomogram.Anatomogram('gallus_gallus')
anatomogram.set_tissue_style('heart', fill='red')
anatomogram.set_tissue_style('lung', fill='blue')
anatomogram.to_matplotlib()

A full list of organs that can be highlighted in the gallus gallus svg image can be obtained by:

import pyanatomogram
anatomogram = pyanatomogram.Anatomogram('gallus_gallus')
anatomogram.get_tissue_names()

Highlight multiple organs with measurement using a matplotlib colormap

Multiple organs can be highlighted by passing a dictionary or a pandas Series to hightlight_tissues. Values will be mapped to colors using a matplotlib colormap.

import pyanatomogram
anatomogram = pyanatomogram.Anatomogram('gallus_gallus')
anatomogram.highlight_tissues({'heart': 0, 'lung': 2, 'brain': 3, 'colon': 1, 'liver': 4}, cmap='Reds')
anatomogram.to_matplotlib()

Style organs with CSS

Custom CSS style attributes can be used to tailor the visualization.

import pyanatomogram
anatomogram = pyanatomogram.Anatomogram('gallus_gallus')
anatomogram.set_tissue_style('heart', **{'fill': 'red', 'stroke': 'black', 'stroke-width': 1, 'fill-opacity': 0.8})
anatomogram.set_tissue_style('lung', **{'fill': 'cornflowerblue', 'stroke': 'darkblue', 'stroke-width': 1,
                                        'fill-opacity': 0.5})
anatomogram.to_matplotlib()

Import into matplotlib axes

The anatomogram can be imported into a matplotlib figure with the to_matplotlib() method, and combined with other figure elements in matplotlib.

import matplotlib.pyplot
import pandas
import pyanatomogram

data = pandas.Series({'heart': 0, 'lung': 2, 'brain': 3, 'colon': 1, 'liver': 4})

fig, axes = matplotlib.pyplot.subplots(nrows=1, ncols=2)

data.plot(kind='barh', ax=axes[0])
axes[0].set_xlabel('Relative expression')
axes[0].set_ylabel('Organ')

cmap = matplotlib.cm.Reds
norm = matplotlib.colors.Normalize()
norm.autoscale(data)
anatomogram = pyanatomogram.Anatomogram('gallus_gallus')
anatomogram.highlight_tissues(data.to_dict(), cmap=cmap, norm=norm)
anatomogram.to_matplotlib(ax=axes[1])

matplotlib.pyplot.colorbar(matplotlib.cm.ScalarMappable(norm=norm, cmap=cmap), ax=axes[1],
                           orientation='horizontal', label='Relative expression')

helper for use in seaborn facetgrid

PyAnatomogram comes with a helper function facetgrid_helper() to be used together with the seaborn FacetGrid class to generate multiple anatomogram subplots faceted by grouping variable(s).

import pandas
import pyanatomogram
import seaborn
import matplotlib

data = pandas.DataFrame({
    'Condition1': {'heart': 0, 'lung': 2, 'brain': 3, 'colon': 1, 'liver': 4},
    'Condition2': {'heart': 3, 'lung': 4, 'brain': 2, 'colon': 2, 'liver': 3},
    'Condition3': {'heart': 7, 'lung': 9, 'brain': 1, 'colon': 7, 'liver': 2},
}).stack().rename_axis(['organ', 'condition']).to_frame('expression')

g = seaborn.FacetGrid(data.reset_index(), col='condition')
cmap = matplotlib.cm.Reds
norm = matplotlib.colors.Normalize()
norm.autoscale(data)
g.map(pyanatomogram.facetgrid_helper, 'organ', 'expression', name='gallus_gallus', cmap=cmap, norm=norm)
g.fig.colorbar(matplotlib.cm.ScalarMappable(norm=norm, cmap=cmap), label='Relative expression')

Generate SVG file

Alternatively to importing the anatomogram into a matplotlib figure, the anatomogram can be saved directly to svg with the save_svg() method.

import pyanatomogram
anatomogram = pyanatomogram.Anatomogram('gallus_gallus')
anatomogram.set_tissue_style('heart', fill='red')
anatomogram.set_tissue_style('lung', fill='blue')
anatomogram.save_svg('anatomogram.svg')