Skip to content

Welcome to OME-Zarr Converters Tools

OME-Zarr Converters Tools is a Python package that provides tooling for building OME-Zarr converters.

It includes three main components:

  1. Abstraction layer for mapping the on-disk raw data to Image objects
  2. Common tooling to build converters as Fractal Compound Tasks

Main Concepts

In general a single microscopy image is not acquired in a single big array in a single file, but rather in multiple smaller tiles. How atomic these tiles are depends on the specific microscope and the acquisition settings.

To make building converters easier, OME-Zarr Converters Tools provides an abstraction layer that allows you to map these on-disk raw data to an Image object which we call Tile.

Moreover, usually a single microscopy image is not composed of a single tile, but rather multiple tiles that are stitched together to form a complete image. We call these objects TiledImage.

flowchart LR
    subgraph A[Metadata Parsing]
    A100[img_B3_fov1_c0_z0.tif] --> B1[Tile1]
    A101[img_B3_fov1_c0_z1.tif] --> B1
    A200[img_B3_fov2_c0_z0.tif] --> B2[Tile2]
    A201[img_B3_fov2_c0_z1.tif] --> B2
    A20x[img_...] --> B3[Tile...]

    B1 --> C1[TiledImage1]
    B2 --> C1

    B3 --> C2[TiledImage2]
    end

    C1 --> D[Init - Task]
    C2 -->|"Many..."| D
    D --> E[Compute Tile1]
    D --> E1[Compute Tile2]
    D -->|Many...| E2[Compute ...]

Additional OME-Zarr Converters Tools supports high-content screening HCS applications. In the context of HCS it is common to have multiple images that are related to each other in a single plate collection. Plates are standardized in OME-Zarr and OME-Zarr Converters Tools provides the necessary tools to correctly place the images in a plate collection.

Installation

To get started with OME-Zarr Converters Tools, you can install it via pip:

pip install ome-zarr-converters-tools