OMERO Virtualization

OMERO (Open Microscopy Environment Remote Objects) is a widely used platform for managing bioimaging data. Its PostgreSQL database stores rich metadata about images, experiments, and annotations — making it an ideal candidate for virtualization with Ontop.

This chapter applies the same OBDA mapping approach from the solar system exercise to a real-world OMERO installation.

Overview¶

The OMERO data model includes tables such as:

• image — microscopy images with acquisition metadata

• experimenter — users/researchers

• experimentergroup — research groups

• screen, plate, well, plateacquisition — high-content screening data

• roi, channel, pixels — image regions and channels

• annotation_mapvalue — key-value annotations attached to images

The full OMERO data model is documented at: OME Model Overview

Prerequisites¶

• Ontop CLI installed (see Installation)

• PostgreSQL JDBC driver in your Ontop jdbc/ directory

• Access to an OMERO PostgreSQL database (the tutorial provides a shared test instance, or you can use your own)

Step 1 — Clone the OMERO mappings repository¶

The omero-ontop-mappings repository contains starter OBDA mappings for OMERO:

git clone https://github.com/German-BioImaging/omero-ontop-mappings -b swat4hcls2026
cd omero-ontop-mappings

Install Ontop CLI (if not already installed)¶

The repo includes a convenience script:

bash utils/install_ontop.sh

Verify the installation:

ontop-cli/ontop help

Step 2 — Explore the OMERO database¶

Useful SQL queries for understanding the database while writing mappings:

List all tables¶

SELECT table_name
FROM information_schema.tables
WHERE NOT starts_with(table_name, 'pg_');

List columns in a table¶

SELECT column_name
FROM information_schema.columns
WHERE table_name = 'image';

The image table contains columns such as:

Step 3 — Launch the Ontop endpoint¶

The repository contains pre-configured artifacts in the omeswat/ directory:

cd omeswat
./omeswat-ontop-endpoint.sh --dev

The log should end with:

-INFO  in i.u.i.o.e.OntopEndpointApplication - Started OntopEndpointApplication

Navigate to http://localhost:8080 to access the Ontop SPARQL query interface.

Step 4 — Explore the existing mappings¶

The mappings in omeswat/omeswat.obda provide a minimal starting point. The Ontop query UI includes several tabs with pre-written queries:

• Playground — free-form SPARQL queries

• Basic Queries — simple retrieval queries

• Tests — ASK queries that validate your mappings

Run the test queries to verify the endpoint is working. All ASK queries should return true.

Assignments¶

The provided mappings are intentionally minimal. Extend them with the following tasks:

Assignment 1: Add entity mappings¶

Create mappings (similar to the image-id mapping) for:

• experimenter

• experimentergroup

• screen

• well

• plate

• plateacquisition

• roi

• channel

• pixels

Each mapping should:

1. Define a class for the entity (e.g. :Experimenter a owl:Class)

2. Map the id and name columns (where available)

3. Add the class and property declarations to the ontology

Assignment 2: Map annotations¶

Create a mapping that exposes image map annotations. OMERO stores key-value annotations in the annotation_mapvalue table. Your mapping should:

• Create blank nodes for each annotation entry

• Add :key and :value properties

• Link the annotation to its parent image

Assignment 3: Write your own mapping¶

Choose an aspect of the OMERO data model that interests you and create a new mapping from scratch. Consider:

• Linking images to their pixel dimensions

• Mapping ROI (Region of Interest) coordinates

• Exposing instrument and objective metadata

Connection to the solar system exercise¶

Notice how the same three Ontop artifacts appear in both use cases:

The pattern is always the same:

SQL database  →  .properties (connection)
              →  .obda (mappings)
              →  .ttl (ontology)
              →  Ontop endpoint  →  SPARQL

Whether you are mapping 8 planets or thousands of microscopy images, the workflow is identical. Only the SQL queries and RDF templates change.