ez_zarr

Goals

The aim of ez_zarr is to provide easy, high-level access to OME-Zarr filesets (high content screening microscopy data, stored according to the NGFF specifications in OME-Zarr with additional metadata fields, for example the ones generated by the Fractal platform).

The goal is that users can write simple scripts working with plates, wells and fields of view, without having to understand how these are represented within an OME-Zarr fileset.

In addition to the python package, we also provide an R package that automatically generates and wraps a python environment with ez_zarr and all dependencies, available at https://github.com/fmicompbio/ezzarr.

Example

You can use ez_zarr from the command line to get information about an OME-Zarr fileset:

ez_zarr tests/example_data/plate_ones.zarr

or from within python to get access to all its functionality:

## import module
from ez_zarr import ome_zarr

## open an Image
img = ome_zarr.Image('tests/example_data/plate_ones_mip.zarr/B/03/0')
img
# Image 0
#   path: tests/example_data/plate_ones_mip.zarr/B/03/0
#   n_channels: 2 (some-label-1, some-label-2)
#   n_pyramid_levels: 3
#   pyramid_zyx_scalefactor: [1. 2. 2.]
#   full_resolution_zyx_spacing (micrometer): [1.0, 0.1625, 0.1625]
#   segmentations: organoids
#   tables (measurements): FOV_ROI_table

## legacy objects from `hcs_wrappers`
from ez_zarr import hcs_wrappers

plate_3d = hcs_wrappers.FractalZarr('tests/example_data/plate_ones.zarr')
plate_3d
# FractalZarr plate_ones.zarr
#   path: tests/example_data/plate_ones.zarr
#   n_wells: 1
#   n_channels: 2 (some-label-1, some-label-2)
#   n_pyramid_levels: 3
#   pyramid_zyx_scalefactor: {'0': array([1. 2. 2.])}
#   full_resolution_zyx_spacing: [1.0, 0.1625, 0.1625]
#   segmentations: 
#   tables (measurements): FOV_ROI_table

A more extensive example is available from here, also available as an ipynb notebook.

Install

Using pip

The release version of ez_zarr can be installed using pip:

pip install ez-zarr

The current (development) ez_zarr can be installed from github.com using:

pip install git+ssh://[email protected]/fmicompbio/ez_zarr.git

Using conda

Alternatively, you can install ez-zarr from the conda-forge channel using:

conda install -c conda-forge --override-channels ez-zarr

Software status

Contributors and License

ez_zarr is released under the MIT License, and the copyright is with the Friedrich Miescher Insitute for Biomedical Research (see LICENSE).

ez_zarr is being developed at the Friedrich Miescher Institute for Biomedical Research by @silvbarb, @csoneson and @mbstadler.