25 Years of Molecular Biology Databases: A Study of Proliferation, Impact, and Maintenance

• Main Author: Heidi J. Imker
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2018-05-01
• Series: Frontiers in Research Metrics and Analytics
• Subjects: databases; research infrastructure; sustainability; data sharing; molecular biology; bioinformatics
• Online Access: https://www.frontiersin.org/article/10.3389/frma.2018.00018/full

Online resources enable unfettered access to and analysis of scientific data and are considered crucial for the advancement of modern science. Despite the clear power of online data resources, including web-available databases, proliferation can be problematic due to challenges in sustainability and long-term persistence. As areas of research become increasingly dependent on access to collections of data, an understanding of the scientific community's capacity to develop and maintain such resources is needed. The advent of the Internet coincided with expanding adoption of database technologies in the early 1990s, and the molecular biology community was at the forefront of using online databases to broadly disseminate data. The journal Nucleic Acids Research has long published articles dedicated to the description of online databases, as either debut or update articles. Snapshots throughout the entire history of online databases can be found in the pages of Nucleic Acids Research's “Database Issue.” Given the prominence of the Database Issue in the molecular biology and bioinformatics communities and the relative rarity of consistent historical documentation, database articles published in Database Issues provide a particularly unique opportunity for longitudinal analysis. To take advantage of this opportunity, the study presented here first identifies each unique database described in 3055 Nucleic Acids Research Database Issue articles published between 1991 and 2016 to gather a rich dataset of databases debuted during this time frame, regardless of current availability. In total, 1,727 unique databases were identified and associated descriptive statistics were gathered for each, including year debuted in a Database Issue and the number of all associated Database Issue publications and accompanying citation counts. Additionally, each database identified was assessed for current availability through testing of all associated URLs published. Finally, to assess maintenance, database websites were inspected to determine the last recorded update. The resulting work allows for an examination of the overall historical trends, such as the rate of database proliferation and attrition as well as an evaluation of citation metrics and on-going database maintenance.