There are hundreds and probably thousands of Web sites that can now be accessed through the Internet for browsing, querying, or retrieving molecular biology data and related data sets. The prospect of (virtual or actual) standard, centralized and non-redundant views of molecular biological data continues to be elusive because of variable levels of reliability of the primary data, the dynamic nature of the databases where the data are stored, and widely disparate database implementation strategies and schema (1). However, their global availability over the Web as well as the ease of accessing them using locally installed Web browsers have made visual browsing and retrieval relatively straightforward. In addition, many of the databases offer query engines serving their individual databases, often using Web-based interfaces.

MBDL (Molecular Biology Database List) includes articles in the annual issue of Nucleic Acids Research devoted to descriptions of sequence and related molecular biology databases. Beyond the articles in the current issue, references to databases described in the previous annual compendia are also included. Additional databases have been included for one or more of several reasons: either those responsible for the database submitted the associated information to us for inclusion, or the database represented a large, centralized resource for the community, or the database represented a data domain not well represented otherwise.

Updates of, or additions to, the current listing are welcome and should be addressed to Christian Burks (burks@exelixis.com). Questions about publishing articles or other aspects of the Nucleic Acids Research special Database Issue should be addressed to Rich Roberts (roberts@neb.com).

Table 1 provides a listing of the initial set of databases that have been included in MBDL, and categorizes them according to a set of relatively arbitrary but useful sub-headings. It is worth noting that, though the primary sequence databases (i.e., protein, nucleotide and polysaccharide) have clearly converged on a few centralized and relatively global resources, the mutations and polymorphisms constituting the second, orthogonal axis of the genome (sequence variation) are still oriented around a distributed, research topic-focused model. Resources have begun to appear presenting comprehensive characterization of RNA and protein as a function of tissue and stage of development, increasingly enabled by the technologies supporting expression arrays and proteomics. Regulatory, signal and metabolic pathways representing the transitive interactions among macromolecules are also increasingly visible on the Web. MBDL provides links from each database listed to its source URL, and also a link to the on-line version of the corresponding article for those described in detail in this issue.