Giffen et al. (2015) demonstrate that promotion, enabling access to biosamples and data, and widening the audience of interest all enhance the value of a historical biobank collection.1 Describing the National Heart, Lung, and Blood Institute (NHLBI) Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC) initiative, the authors explain the steps taken to consolidate the data reserves that led to this finding.
Established in 1976 under the National Institutes of Health (NIH), the NHLBI acted initially as a repository for blood product safety surveys and clinical/epidemiological studies on heart, lung and blood disorders. In addition to holding biosamples, the biobank grew to include data from NHLBI research in 2000. Administrators soon realized that with the growth in the collection, accessing biosamples and allied data could become cumbersome. With this in mind, biobankers instituted a digital database initiative, BioLINCC, that would bring together biosample management with data aggregation under single Web site access for both research requests and repository administration.
The development team started on BioLINCC in 2008 with the aim of maximizing scientific value by increasing access to the various study samples, in addition to promoting the historical collections among a wider research community. The developers used Python v2.7 with Django implementation to build the framework. They used a Centos LINUX operating system v6.6 on an Apache web server v2.4.12 holding a PostgreSQL database v9.1., launching the first implementation in October 2009.
The developers faced three main task areas:
- Organize data and biospecimen inventories
- Develop a Web site portal for users to browse collections and make requests
- Develop standardized workflows to manage requests for biosamples and data and for a review process that oversees the requests
In addition to conducting a thorough physical review of biosamples within the collection to ensure labeling, location and quality standards, the team also worked with staff from these original studies. From this, the biobank ensured that they included additional data that would be useful for onward biobank management and research. The team was then able to link biosamples with corresponding data in addition to ensuring maintenance of patient privacy and confidentiality.
From the review of existing biosamples, the BioLINCC development team created unique searchable Web pages for each existing study. They also converted old pre-digital data records into fully accessible Web content, paying close attention to current Web standards for all users. As an additional bonus, these pages also hold information on building collections for future inclusion in the biorepository.
Increasing the accessibility of the collections to a wider research community could theoretically generate more user requests. This meant that the NHBLI biobank management team needed to streamline the request review process. Working with input from similar institutions, they came up with a hierarchy for classifying collections within the NHBLI repository based on sample abundance and availability. From this review, they classified collections as low, medium or high impact depending on uniqueness and abundance, which indicated how many further study requests could be supported. This classification also influenced the review procedure required for each research request for sample access.
Another classification depended on availability, with studies described as open and freely available, proprietary with biosamples still under the control of the parent study investigators, or non-BioLINCC for collections not administered under the NHBLI biobank.
When first established, BioLINCC covered 8 collections with data from 72 studies. After four years of operation, this rose to managing 38 collections with data from 108 studies, with almost 2,600 registered users of the site. Furthermore, BioLINCC has overseen a rising number of research requests in addition to an increase in user registrations per year. The system has approved requests for access to data collections and teaching data sets, in addition to managing biosample distribution comprising 147,388 specimens arising from 154 requests.
Giffen et al. consider that BioLINCC has been a success, as shown by increased utilization and requests for data and sample access since the program’s launch.
1. Giffen, C.A., et al. (2015) “Providing contemporary access to historical biospecimen collections: Development of the NHLBI Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC),” Biopreservation and Biobanking 13 (pp. 271–279). doi: 10.1089/bio.2014.0050