Data mining and analysis

Stefan Wiemann, Alexander Mehrle, Heiko Rosenfelder

By integrating diverse data resources using suitable analysis tools, the disease driven as well as in the functional genomics and proteomics projects are put into relation.

• Experimental data can be connected with publicly available information, e. g . to map commonly-used biological identifiers (e.g. RefSeq IDs, gene symbols) to the cDNAs investigated thus assigning gene and protein annotation information to in-house data (Mehrle et al., 2006).
• Expression profiling experiments provide information about potential differential expression patterns of genes in cancer tissues. This serves as another criterion for the selection of primary candidates that enter the functional assay pipeline (Arlt et al., 2005).
• Proteins are put into functional and regulatory pathways based on knowledge assembled from external databases, including our experimental information. This step is essential in the validation process, especially to create disease relevant insights.