How can researchers study complex diseases across Europe when sensitive and health data cannot leave hospitals or research centres and speaks several languages?
This is one of the key challenges HEREDITARY is addressing, and part of the answer lies in a powerful combination of a federated learning infrastructure, semantic integration and federated analytics.
In this new #DeCoding article, we take a closer look at how Work Package 3 (WP3) is building the foundations that make this possible: the Hereditary Ontology (HERO) and the Hereditary Data Network (HDN).
From federated learning to federated analytics
In a previous #DeCoding article, we explored federated learning, a method that allows AI models to be trained across multiple institutions without centralising raw data. But before researchers can analyse data across institutions, they need to ensure they are actually talking about the same things. In healthcare, the same clinical concept can be recorded differently depending on the hospital, the specialty, or even the country. This makes it difficult to combine or compare data.
To solve this, HEREDITARY has developed the Hereditary Ontology (HERO): a shared semantic layer that provides a common language for all partners. This allows different datasets to be understood in a consistent way. It enables researchers to formulate questions without needing to know local database structures, by integrating clinical, genomic and imaging data into a unified conceptual model, covering key neurological diseases domains, such as Amyotrophic lateral sclerosis (ALS) and Multiple sclerosis (MS), and designed to expand to others like Parkinson’s and Alzheimer’s.
This semantic integration is essential: without it, federated analytics wouldn’t be possible.
From data silos to a connected network
Building on this ontology, HEREDITARY has developed the Hereditary Data Network (HDN), a federated infrastructure that allows data to be analysed across institutions while remaining locally stored. Data stays where it is, but knowledge can travel. Instead of moving individual patient data to a central repository, HDN enables researchers to send queries to different institutions and receive aggregated results. It is based on a central component that coordinates queries, local endpoints at each institution that execute them on their own data and results that are returned and combined, without exposing sensitive information.
This approach represents a fully federated and privacy-by-design architecture. Privacy controls are integrated in the query processing layer of HDN:
- Each query is assessed before running and it is automatically assigned with a privacy risk score.
- Each institution decides what are the risk thresholds they can safely handle.
- If a query exceeds that thresholds, no data is returned or privacy mitigation measures are applied.
This ensures that data owners remain in full control, while still enabling meaningful research across institutions.
How federated analytics works in practice?
A researcher might ask a question like: “What is the average age at onset of ALS patients?”.
Instead of accessing a central database, the system:
- Translates the question into a standardised query using HERO.
- Sends it to multiple institutions.
- Executes it locally at each site.
- Returns aggregated results.
- Combines them into a single answer, obtaining a response that incorporate insights across different datasets while respecting privacy and institutional autonomy.
Progress so far and what comes next
By now, HEREDITARY has already made significant progress. The project has delivered the first version of its federated workflow execution methods (D3.2) and demonstrated how semantic integration and federated querying can work together. Also, the HDN prototype has shown that distributed queries can be executed across heterogeneous datasets, integrating privacy-aware query mechanisms. For those with a technical interest, various resources relating to these developments can be found on the project’s Open Hub.
Looking ahead, the project is focusing on scaling and real-world deployment. Over the first half of 2026, HDN endpoints are being installed across several partners (University of Turin, Radboud University Medical Centre and University of Colorado), enabling future live queries on real datasets. The goal is to have a fully operational federated query system running at consortium level by the end of 2026, along with a shared catalogue of queries and a clear maintenance plan.
Ultimately, what HEREDITARY is building goes beyond technology. It is a new way of doing research in several fields: one where data does not need to move to generate knowledge, where institutions can collaborate without losing control and privacy, and where complexity is managed through shared understanding. The Federated analytics layer, powered by HERO and the HDN, is a key step in that direction.
Learn more about Federated Analytics in the following videos, where our coordinator, Gianmaria Silvello (University of Padova) and Daniele Dell’Aglio (Aalborg University) share their insights and perspectives on the topic:
Recent Comments