omeROcrate

Integration layer between the OMERO image platform and RO-Crate metadata standard.

Installation

These instructions assume you're using uv.

Create a project if you haven't already:

uv init

You may optionally want to install Glencoe's prebuilt binaries, e.g.

uv pip install https://github.com/glencoesoftware/zeroc-ice-py-linux-x86_64/releases/download/20240202/zeroc_ice-3.6.5-cp39-cp39-manylinux_2_28_x86_64.whl

Then install using:

uv add git+https://github.com/WEHI-SODA-Hub/OmeroCrate

Authentication

You will need to set the following environment variables:

export OMERO_PASSWORD=xxx
export OMERO_HOST=xxx

There are two other optional variables:

OMERO_USER defaults to the current username
OMERO_PORT defaults to 4064

Simple Example

With all the setup done, we can upload some data to OMERO.

In this example we will use the Calcium Imaging data thanks to:

Schröder, M., Staehlke, S., Groth, P., Nebe, J. B., Spors, S., & Krüger, F. (2022). Structure-based knowledge acquisition from electronic lab notebooks for research data provenance documentation. Journal of Biomedical Semantics, 13. https://doi.org/10.1186/s13326-021-00257-x

But you can use your own data instead.

First we obtain the data:

git clone https://github.com/SFB-ELAINE/Ca-imaging-RO-Crate.git

Then we do a simple upload:

uv run omerocrate upload Ca-imaging-RO-Crate/

omeROcrate can do basic crate uploads without configuration.

Advanced Example

RO-Crates are very flexible, which means that you probably use some extra types and properties that aren't handled by omeROcrate by default. Not to worry, you can quite easily customize it to suit your crates, via subclassing.

For example, in the calcium imaging data, there are some entities that look a bit like this:

{
    "@id": "Data/02_Bild-nach-Stimulation_5V_7.9Hz.jpg",
    "@type": "File",
    "https://schema.org/dateModified": {
        "@type": "xsd:dateTime",
        "@value": "2021-03-03T16:58:35"
    }
}

When customizing omeROcrate, it is best to first look at OMERO's own metadata schema and decide if your metadata fits into one of the existing parts of the schema. If not, the metadata can always become an annotation.

In this case, dateModified could be interpreted as the acquisition date of the image. If this is true, we could subclass OmeroUploader to handle this properly. For an example of this, refer to: https://github.com/WEHI-SODA-Hub/OmeroCrate/blob/no-xml/test/calcium_uploader.py

Currently if you subclass OmeroUploader, you will have to perform the upload in Python:

uploader = CalciumUploader(
    conn=connection,
    crate=ca_imaging_1021
)
dataset = uploader.execute()