This project is no longer active and this site is provided for reference only. Please contact grants@elixir-europe.org for enquiries about the project.
Use case: Integrating disease related data and terminology from samples of different types | BioMedBridges

WP10 - Use case: Integrating disease related data and terminology from samples of different types

Activity type: 
RTD
Work package leader: 
Objectives: 

We will demonstrate the feasibility and provide a prototype for linking diseases to molecular information on two levels, terminology and data. We will link:

Both tasks are related since the respective terminologies have to be mapped in order to link biobanks to biomolecular data in ELIXIR databases.

A prototype for an interoperable scheme will link ICD10 terms to genes and protein complexes. This will enable linking biobank and other phenotypic healthcare sector data on individuals to individual genotypes. Interoperability schemes of this kind will be essential for linking BBMRI and ECRIN data to the molecular level and for linking biobank and other phenotypic healthcare sector data on individuals to individual genotypes.

To accomplish the second task, we will select a small number of BBMRI biobanks with the most advanced sample representation in databases. We will develop a model for linking sample objects to the respective objects at the BioSample database developed at EBI. The final deliverable in this work package will be implementation of these links, allowing the user to navigate from the selected biobanks to the EBI BioSample database and vice versa.

The work package will build upon standards developed in WP3, will utilize infrastructure build in WP4 and will benefit from the data security framework developed in WP5.

Description: 

Task 1. Map data elements describing sample information in selected biobank databases that participate in the BBMRI federated infrastructure and the BioSamples database at EMBL-EBI. This work will be done jointly with WP3 to generalize the defined mappings and to develop the minimum standard that would enable exchange of this information.

Task 2. Create a prototype that maps existing healthcare sector terminologies and their phenotypic descriptions to existing repositories and thus links diseases, symptoms and genes. More generally, the aim is to embed the prototype efficiently into existing computational linguistics, bioinformatics and clinical environments alike. In particular, we will link phenotypic terminology in healthcare sector data to genes and we will create a prototype for an interoperable scheme linking ICD9 an ICD10 to gene identifiers in BioSamples, concentrating on genes that have been associated with specific diseases, symptoms and tissues. The prototype will focus specifically on the generic ICD concepts and their mapping to relevant genetic information and will not include efforts that aim to assign codes and terminology to free text in healthcare sector data as it usually is done by text mining. The work on the prototype will also include language interoperability aspects on the terminology side while considering only gene names as they are used in English.

Task 3. Demonstrate the interoperability of the sample representation in BioSamples with the biobank databases participating in the federated biobank information infrastructure created under BBMRI. This will also build upon the work done on terminology and identifier mapping. Working together with WP4 and WP5, we will implement a pilot for federated queries and secure links between a limited number of selected resources in BBMRI and BioSamples. A query system will be developed that will allow sample property based queries across these resources.

Start month: 
13
End month: 
36

Partners

Documents

No documents available to view for this workpackage