For financial and ethical reasons, the large-scale radiobiological animal studies conducted over the past 50 years are, to a large extent, unrepeatable experiments. It is therefore important to retain the primary data from these experiments to allow reanalysis, reinterpretation and re-evaluation of results from, for example, carcinogenicity studies, in the light of new knowledge in radiation biology. Consequently, there is an imperative need to keep these data available for the research community. The European Radiobiological Archives (ERA) were developed to fulfill this task. ERA has become a unique archive, including information from almost all European long-term studies carried out between the 1960s and the 1990s. The legacy database was originally developed in a manner that precluded online use. Therefore, strong efforts were made to transform it into a version that is available online through the web. This went together with quality assurance measures, including first the estimation of the rate of non-systematic errors in data entry, which at 2% proved to be very low. Second, every data set was compared against two external sources of information. Standardization of terminology and histopathology is a prerequisite for meaningful comparison of data across studies and analysis of potential carcinogenic effects. Standardization is particularly critical for the construction of a database that includes data from different studies evaluated by pathologists in different laboratories. A harmonized pathology nomenclature with modern standard pathology terms was introduced. As far as possible, references for the various studies were directly linked to the studies themselves. Further, a direct link to the JANUS database was established. ERA is now in a position where it has the potential to become a worldwide radiobiological research tool. ERA can be accessed at no cost at https://era.bfs.de. An ID and password can be obtained from the curators at era@bfs.de .

ERA has the potential of becoming a world-wide radiobiological research tool for numerous applications, such as the re-analysis of existing data with new approaches in the light of new hypotheses and techniques, and using the database as an information resource for planning future animal studies. When the database is opened for new data it may be possible to offer long-term storage of data from recent and future animal studies.

Molecular and genetic research is now providing us with the opportunity to quantify the risks of radiation exposure at the individual level. The retrospective analysis of earlier epidemiological and animal studies will be an important resource for modelling and evaluating such new risk parameters. With great foresight the EU and EULEP have created a database collecting and collating data from almost all of the available animal radiation biology studies carried out in Europe, the US and Japan between 1960 and 1998, plus those of two human cohort studies. This database is called the European Radiobiology Archives (ERA), which is part of the International Radiobiological Archives (IRA). The archives will be accessible on line via internet. The respective data base is named e.ERA. In addition, e.ERA was made compatible to and interoperable with other relevant radiobiological databases, namely PATHBASE and other mouse phenotype projects. To do so, pathology data in ERA were recoded to conform with accepted ontology standards (MPATH) for the mouse. The data included in the database underwent and undergo continuously quality assurance measures to reassure that the available information is as good as possible. e.ERA has become a unique repository of archival data, the project will soon put into effect an easy to use database of invaluable data for further exploitation by the scientific community. Further, there is the possibility of expanding the database to include data from other projects. During the meeting, the current version of e.ERA will be demonstrated. The interface provides a quick and easy to use way of data retrieval. Data can be browsed by organism, pathology observation, treatment, and various disease codings, e.g. MA, MPath, DisRod, DisFam, ICD9. The complete database version will presumably be available to the scientific community by 2009.