An integrated suite of applications

Common data models and interfaces have been implemented in OBiBa to enable seamless and secure data transfer from one OBiBa application to another. As a result, OBiBa provides studies with an integrated software infrastructure that supports almost all their data management activities and more.

This OBiBa infrastructure can be used for both individual studies and study consortia.

Individual Study

An epidemiological study can take advantage of the whole OBiBa's stack.

Data Collection

Onyx can be used to automate data capture (questionnaire answers, sound recordings, instrument outputs etc.) and thus ensuring a low level of errors (data loss and inconcistencies). Onyx can automatically and securely export these data daily. The data dictionary is always associated to the data export, ensuring that no information will be lost when interpreting the collected data. The export media can be of different type to allow interfacing with various data processing systems:

  • XML files (optionally encrypted),
  • CSV files,
  • Direct secured connection to a Opal server.

Data Storage

Opal is able to store data from Onyx but also from other data sources, usually CSV or SPSS files.

The data dictionary can be prepared before or after the data importation. Data can also be transformed by developing derivation algorithms for data cleaning and quality checks. The participants' data can also be enriched with other type of information: drug registry, geographical area metrics etc.

The specificity of Opal is also that participant identifiers can be compartmentalized for a better privacy and to avoid data collusion: the same participant can have one ID in the data collected, another one in Opal storage and another one again in the data exported to a research group.

Data Analysis

Data stored in Opal can be exported to files in common data formats or can be pushed on-demand into R for performing statistical analysis, graphics and reports.

Data Publication

Mica is the web portal of choice for publishing data dictionaries and aggregated results. Out-of-the-box search capabilities of Mica allow researcher to explore study's variables and data profiles that could match their project needs. Data access requests can also be submitted and evaluated.

Study Consortia

Obiba offers a unique software solution to support data collection, harmonization and federation. As depicted in the figure below:

  • Data collected from Onyx (or from other sources) can be imported into study-specific Opal databases,
  • Each study then uses the Opal’s ability of creating views and derived variables to implement processing algorithms that transform data collected by each study into a common (i.e. harmonized) dataset,
  • Making use of Mica and Opal, harmonized data can then be queried and analyzed with the DataSHIELD approach through a federated database system, without sending individual-level study data outside of host institutions.

Real examples of such OBiBa federated and harmonized database is described in CPTP and BioSHaRE user stories.

To set up or evaluate your own federated database infrastructure using OBiBa, follow the Data Harmonization and Database Federation Tutorial .