Will a Biological Database Be Different from a Biological Journal?

The differences, or otherwise, between biological databases and journals is an important question to consider as we ponder the future dissemination and impact of science. If databases and journals remain discrete, our methods of assimilating information will change relatively little in the years to come. On the other hand, if databases and journals become more integrated, the way we do science could change significantly. As both Editor-in-Chief of PLoS Computational Biology and Codirector of the Protein Data Bank (PDB), one of the oldest and widely used data resources in molecular biology, the question is particularly pertinent. Here, I give my perspective on what could and, I believe, should happen in the future. My vision is that a traditional biological journal will become just one part of various biological data resources as the scientific knowledge in published papers is stored and used more like a database. Conversely, the scientific literature will seamlessly provide annotation of records in the biological databases. Imagine reading a description of an active site of a biological molecule in a paper, being able to access immediately the atomic coordinates specifically for that active site, and then using a tool to explore the intricate set of hydrogen-bonding interactions described in the paper. Not only are the data generated by the experiment immediately available within the context of what you are reading, but specific tools for interpreting these data are provided by the journal. Alternatively, if you are starting with the data, for example, viewing the chromosome location of a human single-nucleotide polymorphism associated with a neurological disorder, you can immediately access a variety of papers ranked in order of relevance to your profile, not just through links to abstracts but also by pinpointing the reference to the single-nucleotide polymorphism in the fulltext article. The type and order of articles displayed could be different, depending on whether you are, for example, a molecular biologist or a neurosurgeon. At this point, whatever your user profile, the distinction between a database and a journal article disappears. How could this happen? To answer this question, we must think about the parallels that exist today between biological databases and biological journals. The daily work of any high-throughput scientific journal or biological database consists of information input, information processing, and information output. Consider the parallels between a journal and a database for each of these three steps. On a daily basis, the journal accepts manuscripts; once these have been checked for format compliance and completeness, they undergo review, either by an internal group of scientific editors or, as is the case for PLoS Computational Biology, through peer review by the scientific community. Likewise, a biological database such as the PDB accepts submissions from the community, which are checked for format compliance and reviewed internally by experienced annotators. There are even parallel presubmission steps in journals and databases. For example, potential authors in PLoS Computational Biology may make presubmission inquiries to confirm the suitability of their paper, and depositors to the PDB may run their entries against a validation server to determine whether the data are in compliance, prior to having the same tests run by a PDB annotator. Once registered with the corresponding online submission system, a journal manuscript receives a permanent manuscript number, while a database entry receives a unique identifier. Subsequent revisions can be mapped to these respective numbers, so that both journals and databases can provide an accurate audit trail of journal manuscripts and database entries, respectively. Once a manuscript or entry is accepted as compliant, both undergo review processes involving one or more iterative steps between institution and author, as the manuscript or the entry is refined and finally released. Release cycles of journals and databases have also become similar—journals such PLoS Computational Biology have an option for early online release as soon as the manuscript is accepted, and biological databases typically release entries on a daily or weekly basis, as soon as they have been processed. Not only are the daily operations of databases and journals similar, but the business models also have parallels (I will not dwell on them here though). Certainly from a consumer’s perspective, in terms of accessibility, there is no difference between a paper in a PLoS journal and an entry in the PDB database—they are freely available to all. In the case of openaccess journals and open archives like the PDB, the parallels, from the perspective of the consumer, are even more profound than just free access yet are frequently overlooked. PLoS articles are published under a Creative Commons Attribution License, which means that the contents (text and images) of all PLoS journals can be used as the consumer sees fit, provided original attribution is given to the appropriate authors and source. So it is with the contents of many biological databases, including the PDB. Consumers are free