The Cambridge Structural Database (CSD) is the world’s repository of crystallographic results for organic and metal-organic compounds. The database now stores bibliographic, chemical and numerical data for over 290,000 crystal structures, and it increases by 24,000 new structures each year. The database is heavily used as a research tool in structural chemistry and in life sciences applications, particularly those related to drug discovery. Scientists who make use of the CSD wish to cite the most recent reference to the database, and this is the first major update paper to appear for more than a decade. The paper appears in a journal which is readily accessible, well regarded, and highly appropriate for the subject matter.

The paper provides a major update on the CSD itself, including references to database access software supplied as part of the CSD System and major reviews of CSD research applications, which all appeared in a special issue of Acta Crystallographica (Sections B and D, June 2002). Rather than describing a new discovery or methodology (the CSD has been distributed internationally since the late 1970s), it provides a comprehensive overview of the CSD of today, together with statistics that chart the development of small-molecule crystallography. The related papers, however, do describe new software methodologies, and provide detailed surveys of the types of CSD-based studies that are contained in more than 1,000 research applications published internationally. The whole purpose of CSD System development, and the development of CSD-related software products in structural chemistry and the life sciences, is to make crystal structure data more useful to others.

Crystal structure analysis is now the method of choice for structural characterisation of both known and novel small molecules. Apart from determining atomic connectivity, the technique also provides accurate geometrical data concerning the molecule itself and the interactions it makes with other molecules in the extended crystal structure. Such data underpin systematic studies of substructural features, such as mean molecular dimensions, preferred conformations, structure-energy relationships, etc. Knowledge of the geometries and directionalities of hydrogen bonds and other non-bonded contacts is vital in crystal structure prediction, crystal engineering, and in predicting the docking modes of novel ligands with protein active sites. The CSD is a complete and accurate source of atomic coordinate data which are linked directly to the underlying chemistry to permit substructure searching in both two and three dimensions. The CSD and its associated software allow the user to explore, visualize, and analyze molecular structure and molecular interactions at a highly detailed level.

Following a Ph.D. and postdoctoral research in crystallography, I was looking for a position that coupled crystal structure analysis with the systematic analysis of its results in structural chemistry. In 1970, I joined a group in Cambridge University that had just started (since 1965) to create a database of the known crystal structures of small molecules. The CSD was one of the first numerical scientific databases to be created anywhere, and presented significant challenges of organization using the computer resources then available. As the number of published crystal structures began to rise, I became more and more involved in developing software for database creation and access, helping to process the increasing floods of numbers, and in using the accumulated data in structural chemistry research.

Frank H. Allen, Ph.D.
Executive Director
Cambridge Crystallographic Data Centre
Cambridge, United Kingdom