Global systems for biomedical research, GlobalBioRes, is a multidisciplinary project in computing science and bioinformatics. It will enable automated data integration from web databases and develop new visualizations supporting the understanding of complex biomedical data sets.
Researchers using post-genomic technologies perform large-scale experiments, which can only be interpreted in the light of data harvested from many web databases. Although data harvesting can be automated, automated removal of duplicate values and data merging are not supported, and are not possible with current database technologies.
There are also no adequate visualization tools to offer the combined analysis of textual data sets and the physical framework of genomes and chromosomes. The project will develop and test data mining algorithms on around 15 biological databases available in XML.
As the schemas are dissimilar, data mining will apply to data values indexed in a relational database identify mappings between dissimilar XML subtrees, based on data mining results Record those mappings as tree subsumptions, and index them Develop and test algorithms which use subsumptions to remove duplicated data values and merge the data trees visualize integrated data in the context of genome and synteny maps.
Iteratively test both data integration and visualization approaches, in order to develop new insights on what mechanisms are effective in data integration and visualization. The proposed system will be tested by our biomedical collaborators in disease gene finding, drug target discovery, and drug delivery research. It is hoped to bring immediate efficiency and quality gains.
In the long term this work will contribute to new technologies for e-Science, e-Business and e-Government, and feed into database technologies. The project will allow the applicant to move abroad to complete her postdoctoral training in collaboration with a recognised research leader.
Call for proposal
See other projects for this call