Describing the quality of curated e-Science information resources

Qurator project goals

e-Science involves the sharing of experimental results on a global scale. Scientists now expect to make use of information produced by other labs/projects in validating and interpreting their own results. However, many unsolved problems remain in the area of information sharing, not least of which are those caused by hidden differences in information quality (IQ) between data sources.

This project seeks to assist scientists and data curators in managing the quality of their information. Rather than trying to impose a single approach and set of IQ priorities, an alternative is to provide scientists with the means of annotating their information with explicit descriptions of its quality in terms of locally relevant characteristics.

Working closely with user-scientists in two post-genomics domains - proteomics and transcriptomics - two hypotheses will be tested:

1. that it is possible to elicit detailed specifications of the IQ priorities of specific scientific domains;
2. that the annotation of sources (relative to the identified priorities) can be performed in a cost-effective manner.

The two specific domains have been chosen because scientists in these areas have identified genuine IQ concerns. However, every effort will be made to ensure the generality of the results to other e-Science areas.