Figure 3From: A random set scoring model for prioritization of disease candidate genes using protein complexes and data-mining of GeneRIF, OMIM and PubMed recordsReceiver operating characteristic (ROC) curves of prioritizations using different phenotype sources and vocabulary filters. Each ROC curve represents the prioritization performance when combining a specific gene-associated phenotype with a vocabulary filter. The phenotype sources were OMIM (brown), PubMed (green), and GeneRIF (purple). The vocabulary filters were STY, MeSH, ICD9CM, and GO (colored from dark to light accordingly).Back to article page