Integrative network approach.

<p>Genes with a genome-wide significance of P<10⁻⁰⁸ or less associated with Parkinson's disease (displayed in blue) and type 2 diabetes (displayed in purple) were included in this study and specified as training genes. A random walk algorithm within the functional linkage network (displayed in gray) was performed to identify candidates genes with a linkage distance (LD) to the training genes of less than or equal to 1 within the FLN. Candidate genes (displayed in orange) were ranked and scored according to their closeness with training genes. Data from microarrays studies in blood of Parkinson's disease, pre-diabetes and Type 2 diabetes patients was analyzed to identify genes dysregulated in both diseases. Quantitative PCR assays were used to validate a potential biomarker in blood of Parkinson's disease patients. Networks were visualized using Cytoscape 2.8.3. PD = Parkinson's disease, IFG = pre-diabetes and T2D = type 2 diabetes.