PhenomeNET is a cross-species phenotype network of phenotypic similarity. It currently contains 86,203 complex phenotype nodes taken from the yeast, fish, worm, fly and mouse model organism databases as well as human disease phenotypes from OMIM. The network is a complete graph in which edge weights represent the degree of phenotypic similarity. To compute similarity, we utilize the Jaccard metric weighted by information content of ontology terms.
PhenomeNET is available in two formats:
pre-filtered (similarity score >0.1) as list of node pairs and similarity score (Warning: 2.1GB). The file can be explored using grep and sort or similar tools to investigate associations between nodes in PhenomeNET. The file can be used as input for Cytoscape.
adjacency matrix (upper right half only) (Warning: 11GB). This file contains the column labels of the matrix. The file is intended for programmatic access to the raw data and requires additional software. See KeggEval.groovy, line 287ff, for example code.
If you use PhenomeNet, please cite:
Robert Hoehndorf, Paul N. Schofield and Georgios V. Gkoutos. PhenomeNET: a whole-phenome approach to disease gene discovery. Nucleid Acids Research, 2011.