Figure 2
From: Extraction of semantic biomedical relations from text using conditional random fields
Sample subgraphs of the gene-disease graph. Diseases are shown as squares, genes as circles. The entities for which associations are extracted, are highlighted in yellow. We restricted ourselves to genes, which our model inferred to be directly associated with Parkinson's disease, regardless of the relation type. The size of the nodes reflects the number of edges pointing to/from this node. Note that the connectivity is calculated based on the entire subgraph, whereas (a) shows a subgraph restricted to altered expression relations for Parkinson, Alzheimer and Schizophrenia and (b) shows a genetic variation subgraph for the same diseases.