Welcome to BiobankUniverse

Introduction

BiobankUniverse is a data discovery and matching system for biobanks built in MOLGENIS that can automatically tag biobank attributes (metadata) with ontology terms from Unified Medical Language System (UMLS) and based on which generate candidate attribute matches between biobanks. In addition, BiobankUniverse can calculate the overall similarities between biobanks so that users can quickly get an approximation of how close/far the biobanks are to each other.

The instruction on how to deploy it in tomcat 7 can be found here. The source code is avaiable on GitHub


Discover biobanks

The circle represents a biobank and the larger a biobank, the larger the circle. The line represents the distance between the biobanks, the longer the line, the less similar the connected biobanks are to each other. There are two types of universes, one is generated based on the overall similarity of the automatically tagged ontology terms, the other one is generated based on the number of candidate matches.

For example, click on the button 'Number of matches', type in 'Beer' in the search box and select 'C0004922:Beer' and then hit the 'Search' button, only the candidate matches related to 'Beer' are retrieved and universe will reduce in size. The search box is configured with typeahead function to provide ontology terms, please be patient if the ontology term list doesn't show up fast.




Network


Curate attribute matches

We have pre-generated the candidate attribute matches between the biobanks including Finrisk2002, Finrisk2007, Lifelines, Prevend, Mitchelstown and HOP. The matches generated are pairwise so users need to choose a leading 'target' data schema to generate the match table, in which they can curate the candidate matches.



Technical design

Not available