capDSD - confidence, augmented pathway diffusion state distance

DSD is a program that calculates diffusion state distances between nodes in a graph. It makes use of global topologies via random walks and captures proximity between nodes in terms of node properties such as protein functions among protein-protein interaction graph.

cDSD extends this by including information about interaction confidence

caDSD is build on top of cDSD, and augments it with high-confidence signaling pathway data

capDSD extends caDSD by making use of the topological properties of these signaling pathways

If you use capDSD, please cite:

M. Cao, C. M. Pietras, X. Feng, K. J. Doroschak, T. Schaffner, J. Park, H. Zhang, L. J. Cowen and B. Hescott, New directions for diffusion-based network prediction of protein function: incorporating pathways with confidence Bioinformatics, Volume 30, ISMB 2014 Proceedings, i219-i227, 2014.

capDSD:

capDSD-src.zip.

A small example ppi and set of pathways is available here

Readme (Also included in the source archive)

Raw data used in the paper are as follows:

• Physical PPIs from BioGRID (BioGRID official website is here)


• KGML from KEGG (KEGG official website is here)


• MIPS annotations (Note: the MIPS FunCat mapping between indices and categories is here)
  
  • MIPS First Level Annotation
    
  • MIPS Second Level Annotation
    
  • MIPS Third Level Annotation
    

Intermediate data used in the paper are as follows:

• Physical PPIs from BioGRID with confidence values (using the t100 method detailed in the paper)


• Pathways from KEGG in PPIP format

DSD matrices (Warning: Large)

Other toolkits:

• source code for calculating confidence scores
.
• source code for parsing KEGG pathways (included in the capDSD source archive)