Light Mastodon Reader for Python

This is a lighter version of Mastodon Reader for Python from Chrisoph Sommer.

Import the spots and links tables, features, tags and meta data from a Mastodon project file.

Ported to Python from: matlab-mastodon-importer

This reader is lighter mainly because it does not depend on pandas nor on networkx and it therefore only depends on numpy.

Example

Read the mastodon file:

from micromastodonreader import MicroMastodonReader

mr = MicroMastodonReader("demo/demo.mastodon")

# show meta data
meta_data = mr.read_metadata()

# read spot and link tables with features and tags columns
spots, links, tag_definition, features = mr.read(tags=True, features=True)

spots is a numpy.ndarray where:

• spots[int_id] is the cell with id int_id
• spots[int_id, 0] is the x coordinate of the cell int_id
• spots[int_id, 1] is the y coordinate of the cell int_id
• spots[int_id, 2] is the z coordinate of the cell int_id
• spots[int_id, 3] is the time of the cell int_id
• spots[int_id, 3:10] are the values of the elongation matrix representing the cell int_id sphere
• spots[int_id, 10] is the bsrs (??) of int_id

links is a numpy.ndarray where:

• links[int_pos] is an edge
• links[int_pos, 2] is the id of the edge
• links[int_pos, 0] is the id of the cell at time t
• links[int_pos, 1] is the id of the cell at time t + 1

tag_definition and features are a bit more complicated 🫠

You can also read information separately

# read only spot and link tables
spots, links = mr.read_tables()

# read tag_definition
tag_definition = mr.read_tags()

# read features
mr.read_features()

Installation

Current version

pip install git+git://github.com/GuignardLab/python-mastodon-importer-light

pip

pip install micro-mastodon-reader

more information on PyPi