开源软件名称（OpenSource Name）：

bercowskya/MastodonAnalysis

开源软件地址(OpenSource Url)：

https://github.com/bercowskya/MastodonAnalysis

开源编程语言(OpenSource Language)：

Jupyter Notebook 92.7%

开源软件介绍(OpenSource Introduction)：

Paleontologist

Read and analyse Mastodon (FIJI Plugin) .csv and .xml data

Mastodon is a large-scale tracking and track-editing framework for large, multi-view images. It allows you to track cells' dynamics over time and has a very useful and easy to use GUI. In order to use Mastodon, since it works with Big Data Viewer, you need your data to be in HDF5/xml format.

As an output, Mastodon provides either a -mamut.xml or a .csv file which containes, among many features, the XYZ coordinates of each cell and the 3D average intensity of the cells. This package provides the tools to facilitate the organization of the data and enable the easy creation of figures for spatial, temporal and mitotic dynamics of the cells.

[1] Conversion to HDF5 and XML

Before using Mastodon, you need to convert your files in a format that BigData viewer can read. For this, using either Big Data Viewer, BigStitcher or Multiview Reconstruction from Fiji, you can convert your data into HDF5 and XML. HDF5 will save the raw data whereas the XML file will save the metadata and any transformation performed to the raw data.

[2] Using Mastodon

Mastodon is a very user-friendly Tracking plugin from Fiji. It allows interactive visualization and navigation of large images thanks to the BigDataViewer. Any file that can be opened in the BigDataViewer will work in Mastodon (BDV HDF5 file format, KLB, Keller-Lab Blocks file format, N5 file format, ...).

With Mastodon you will be able to track large amount of cells in a manual, semi-automatic or automatic way. The outputs from the tracking are two .csv files: name-edges.csv and name-vertices.csv . The first one contains the information obtained from the spots: mean, median and standard deviation of intensity of all the channels; x, y, z coordinates of the centroid of the spots; spots radius; detection quality for each spot; tags and sub-tags for the spots; the individual ID for each spot; the track ID to which each spot corresponds.

[3] Using MastodonAnalysis.py to analyze Mastodon data

MastodonAnalysis.py is a collection of classes that allows you to obtain tidy arrays with the tracks' and spots' features in an easy way.

Below you can find a description for each of these classes:

Class `xml_features`:

Gets as input the .xml file from the initial conversion using either BigdataViewer, Bigstitcher or Multiview reconstruction to convert the files into HDF5/XML. This class can be called by using the following line of code:

fts = xml_features(path_xml)

where path_xml has the directory where the path of the .xml and .hdf5 files are. This line of code saves the output of the class xml_features into the object fts. Therefore, if you write fts. and then press Tab you will get all possible outcomes from this class. The list of these outcomes are:

channels
dimensions
width
height
number of slices
x,y,z pixel size
coordinate units ($ \mu $ m, mm, etc.)

Class `csv_features`:

This class allows you to obtain all the information that comes in the .csv files (-vertices and -edges) that are generated with Mastodon once you have computed the features in Mastodon and saved the results in the .csv format. The file called name-vertices.csv contains all the information concerning each individual spot. The file called name-edges.csv contains all the information concerning the links of each spot.

To call this class use the following line of code:

spots = csv_features(path_csv, path_xml)

This line of code saves the output of the class xml_features into the object spots. Therefore, if you write spots. and then press Tab you will get all possible outcomes from this class.

The list of the outcomes are: