Some of the experiences of the CombeChem e-Science project in relation to both automation and the need for semantics in combining modern computer science techniques and chemistry are discussed. In particular the aspects of the smart laboratory, large scale data handling and the way this impacts on the necessary database technology are discussed. In addition some of the ways in which the grid can enable greater user interaction with services such as the National Crystallography Service and improve the dissemination of the resulting data via the e-Bank approach are presented.

eBank Combechem

The paper will illustrate the ?CombeChem Project? experience of supporting the chemical data lifecycle, from inception in the laboratory to organization of the data from the chemical literature. The paper will follow the different parts of the data lifecycle, beginning with a discussion of how the laboratory data could (or should) be recorded, and enriched with appropriate metadata, so as to ensure that curated data can be understood within its original context when subsequently accessed, as it is generated (the ideal of ?Autonomic Annotation@Source?). Intrinsic to our argument is the recording of the context as well as the data, and maintaining access to the data in the most flexible form for potential future re-use for purposes that are not recognised when the data was collected. This is likely to involve many routes to dissemination, with data and ideas being treated by parallel but linked methods, which will influence traditional approaches to publication and dissemination, giving rise to a Grid style access to the information working across several administrative domains summarized by the concept of ?Publication@Source?.

Advances in crystallographic instrumentation and computational resources have caused an explosion of crystallographic data, as shown by the recent exponential growth of the Cambridge Structural Database (CSD) [1]. However, even this is considered to be lower than expected, following the introduction of area detection. The reason for this is clearly identified as a publication bottleneck [2], which can only become even more severe with developments in high throughput crystallography [3]. As a result of this situation, the user community is deprived of valuable information, and the funding bodies are getting a poor return for their investments! Unlike the mathematical and electronic sciences, the chemical sciences have been reluctant to embrace the 'preprint concept' [4]: the one exception has been the efforts of rapid electronic communications journals. This poster outlines a publication at source procedure for the rapid and effective dissemination of structural information to the scientific community which removes the lengthy peer review process that hampers traditional publication routes, but provides an alternative mechanism. eCrystallographyDataReports are built on a concept developed in the Computer Science community, the Open Archive Initiative (OAI) [5], whereby an author may reveal to the public archives of information. OAI software, known as ePrints [6], has been modified to implement this new type of open archive. An eCrystallographyDataReport makes available all raw, derived and results data from a crystallographic experiment via a searchable and hierarchical system. At the top searchable level this metadata includes bibliographic and chemical identifier items which allow access to a secondary level of crystallographic items which are directly linked to the associated archived data. The 'core bibliographic data' is made available to the public domain by the Open Archive Initiative Protocol for Metadata Harvesting (OAI-PMH) [7] so that it may be 'harvested' by data aggregator services. This enables linking of the archive record with other, perhaps not structurally related, published literature. Perhaps more importantly it notifies those monitoring the archive that a new entry is available, hence the CSD may automatically download the structural data and assimilate it into the database. This will return a deposition number which, along with any future literature citations of this entry, can be retrospectively added to the record to link the data or information sources. eCrystallographyDataReports are intended for the publication of data only. Hence the results of a crystal structure determination may be disseminated in a manner that anyone wishing to utilise the information may access the entire archive of data related to it and assess its validity and worth. The archive entry can then be referenced, which allows learned society journals to streamline articles to contain more intellectual discussion and less experimental reports. This archive software is to be freely available for public download and installation and is simple and inexpensive to set up and maintain.

Recent advances in crystallographic instrumentation and computational resources have caused an explosion of crystallographic data, as shown by the recent exponential growth of the CSD [1]. However, even this is considered to be lower than expected, following the introduction of area detection. The reason for this is clearly identified as a publication bottleneck, which will become even more severe with developments in high throughput crystallography [2]. As a result of this situation, the user community is deprived of valuable information, and the funding bodies are getting a poor return for their investments! Electronic publishing has helped to make some inroads into the problem, and is already an integral part of chemical and crystallographic publishing. This has dramatically reduced the turnaround time in the publication process, but the system still fails to keep apace of the generation of structural information, mainly because of the continued reliance on traditional protocols for the assembly of manuscripts and their peer review. The need to maintain some kind of review process is clear, and this makes the sometimes-used direct submission of structures to the CCDC ?unpopular?. Unlike the mathematical and electronic sciences, the chemical sciences have been reluctant to embrace the 'preprint concept' [3]: the one exception has been the efforts of rapid electronic communications journals. This poster outlines a pre-print procedure for the rapid and effective dissemination of structural information to the scientific community which removes the lengthy peer review process that hampers traditional publication routes, but provides an alternative mechanism. Crystallographic EPrints are built on a concept developed in the Computer Science community [4] whereby an author may reveal to the public archives of information. An Eprint makes available all raw, derived and results data from a crystallographic experiment via a searchable and hierarchical system. At the top searchable level this metadata includes bibliographic and chemical identifier items which allow access to a secondary level of searchable crystallographic items which are directly linked to the associated archived data. Hence the results of a crystal structure determination may be disseminated in a manner that anyone wishing to utilise the information may access the entire archive of data related to it and assess its validity and worth.