The Informatics Challenges Facing Biobanks: A Perspective from a United Kingdom Biobanking Network

Abstract

The challenges facing biobanks are changing from simple collections of materials to quality-assured fit-for-purpose clinically annotated samples. As a result, informatics awareness and capabilities of a biobank are now intrinsically related to quality. A biobank may be considered a data repository, in the form of raw data (the unprocessed samples), data surrounding the samples (processing and storage conditions), supplementary data (such as clinical annotations), and an increasing ethical requirement for biobanks to have a mechanism for researchers to return their data. The informatics capabilities of a biobank are no longer simply knowing sample locations; instead the capabilities will become a distinguishing factor in the ability of a biobank to provide appropriate samples. There is an increasing requirement for biobanking systems (whether in-house or commercially sourced) to ensure the informatics systems stay apace with the changes being experienced by the biobanking community. In turn, there is a requirement for the biobanks to have a clear informatics policy and directive that is embedded into the wider decision making process. As an example, the Breast Cancer Campaign Tissue Bank in the UK was a collaboration between four individual and diverse biobanks in the UK, and an informatics platform has been developed to address the challenges of running a distributed network. From developing such a system there are key observations about what can or cannot be achieved by informatics in isolation. This article will highlight some of the lessons learned during this development process.

Architecture Enabling Service-oriented Digital Biobanks

“…This is preceded by a definition of a reliable method to identify researchers and to manage access rights to the services. [22,15] The biobank is to store data indefinitely or until consent for the data is revoked. The structure of the basic data and logic for linking should therefore be simple.…”

“…Finally, the distribution of data and samples to researchers has to be managed. This requires sufficient information capabilities [15] including availability services like catalogues to deliver the data to users. This occurs either through online viewing or other data distribution methods.…”

“…The Qpati system has been in use in several hospitals in Finland and updates on the system have gradually been added, consequently different hospitals may have a different version in use. Different hospitals may also use different clinical terms [15].…”

Building a ‘Repository of Science’: The importance of integrating biobanks within molecular pathology programmes

“…However, the informatics capabilities of modern biobanks must also extend beyond just sample tracking and traceability. [15] Much of the success of a biobank will also depend on their ability to link samples to relevant and robust de-identified clinical, pathological and increasingly complex diagnostic molecular pathology information. This is critical as it allows researchers not only to further their understanding of tumour development and heterogeneity but to track the course of the disease including response to specific treatment regimens and final clinical outcomes.…”

Accelerating the Development and Validation of New Value-Based Diagnostics by Leveraging Biobanks

“…They all rely on state-of-the-art standards in clinical, genomic, and translational data. These need to be built in from the beginning in the operation of biobanks [25]. …”