Additional file 5: Figure S3. of Gene expression and functional annotation of human choroid plexus epithelium failure in Alzheimerâ s disease

Downregulated genes in AD Br5â 6. The input genes Additional file 8: Table S4) are bioinformatically translated to proteins and the symbols are colored. Another set of genes/proteins is inserted by the knowledgedatabase to construct the most likely networks; the symbols are white. Different types of symbols exist (for explanation see www.ingenuity.com : transporters, strucural proteins, secreted proteins, etc). The symbols are connected by solid lines and dotted lines. These denote, respectively, direct physical/functional and indirect interaction between the symbols/genes/proteins. These interactions are, via the knowledgde database, derived form wet lab experiments in the literature, and from big data stored in curated public databases (GEO, etc). The most likely networks are constructed by the knowledge database, taken all available data into account of human, mouse, rat and in vitro model experiments. Finally, once the networks are constructed, specific entries in the networks can be labelled, by hand of the user, to specifically label molecules wihch are involved in sopecific diseases or biological process of interest. In this way, the (predicted) molecular enviroment of entries of interest can be explored. (ZIP 10804 kb)