Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/6E1AE777-8478-4A67-84BD-9AC34FF50D94 | ||
| Year | 2009 | ||
| Type of Item | Conference Full Paper | ||
| License |
|
||
| Bibliographic Citation | E. Koumakis, S. Sfakianakis, V. Moustakis, and G. Potamias. (2009). Discovery of Genotype‐to‐Phenotype Associations: A Grid‐enabled Scientific Workflow Setting. Presented at Biomedical Informatics and Intelligent Methods in the support of Genomic Medicine Workshop – Artificial Intelligence Applications and Innovations International Conference. [Online]. Available: http://www.researchgate.net/publication/220828639_Discovery_of_Genotype-to-Phenotype_Associations_A_Grid-enabled_Scientific_Workflow_Sett | ||
| Appears in Collections |
The heterogeneity and scale of the data generated by high throughput genotyping association studies calls for seamless access to respective distributed data sources. Toward this end the utilization of state of the art data resource management and integration methodologies such as Grid and Web Services is of paramount importance for the realization of efficient and secure knowledge discovery scenarios. In this paper we present a Grid-enabled Genotype to Phenotype scenario (GG2P) realized by a respective scientific workflow. GG2P supports seamless integration of clinico-genetic heterogeneous data sources, and the discovery of indicative and predictive clinico-genetic models. GG2P integrates distributed (publicly available) genotyping databases (ArrayExpress) and utilizes specific data-mining techniques for feature selection - all wrapped around custom made Web Services. GG2P was applied on a whole-genome SNP-genotyping experiment (breast cancer vs. normal/control phenotypes). A set of about 100 discriminant SNPs were induced, and classification performance was very high. The biological relevance of the findings is strongly supported by the relevant literature.
Copyright © DIAS 2013
