You can download the Python package implementing PathLinker from by visiting T. M. Murali's software page. We request you to use this option so that we can store your email address for updates and for record keeping. If you do not prefer to give us your email address, you can download PathLinker directly from GitHub.
The human interactome is a directed graph constructed from a number of sources, as described in the paper. The interactome is a tab-delimited file with four columns: tail node id (UniProtKB), head node id (UniProtKB), edge weight, and evidence for the edge. If multiple evidence types or sources support an edge, a pipe (|) separates. Edges corresponding to physical interactions appear in both directions in the file.
GraphSpace is an interactive graph-sharing website developed by T. M. Murali's group. Signaling pathway reconstructions computed by PathLinker are available as public graphs (no login needed). All these reconsructions may be downloaded from GraphSpace as PNG images and as JSON files (see Interacting with Graphs on the GraphSpace Help Page).
In addition to GraphSpace, PathLinker's reconstructions for the 15 NetPath pathways used in the aggregate analysis are available as tab-delimited files. These files follow the same convention as the human interactome, with an extra column that denotes the paths in which each interaction appears. The top 200 paths of each file are displayed in GraphSpace under the pathlinker-all-netpath-pathways tag.