C-GRAAL: Common-Neighbors-Based Global GRAph ALignment Algorithm
Supplementary material
Vesna Memišević and Nataša Pržulj
Corresponding author: Prof. N. Pržulj, natasa.przulj@mbzuai.ac.ae
Downloads
About
- V. Memišević and N. Pržulj. C-GRAAL: Common-neighbors-based global GRAph ALignment of biological networks. Integrative Biology 4, 734–743 (2012). PMID: 22234340
- Supplementary material (PDF)
Executables
Networks
Alignments
Predictions
- Yeast - Human
- F. tularensis - H. sapiens
- B. anthracis - H. sapiens
- Y. pestis - H. sapiens
- C. jejuni - E. coli
- Mesorhizobium loti - Synechocystis sp. PCC6803
How to run C-GRAAL
- Download
CGRAAL_unix64from the link above. C-GRAAL accepts networks in LEDA.gwformat. - Run
./CGRAAL_unix64without parameters to see usage directions.
Usage:
./CGRAAL_unix64 net1.gw net2.gw similarity.file output1.file output2.file
-
net1.gw: first network (.gwformat) -
net2.gw: second network (.gwformat) -
similarity.file: similarity between nodes in net1 and net2, in the formatnodeXNet1 nodeYNet2 similarity -
output1.file: alignment output asnode_Net1 node_Net2, with nodes as numbers -
output2.file: alignment output asnode_Net1 node_Net2, with nodes as names
The number of nodes in the first network must not be greater than the number of nodes in the second network.
If you get a “permission denied” error, run chmod +x CGRAAL_unix64 first.
Converting an edge list to .gw
If your network is an edge list:
node1 node2
node3 node4
...
convert it with the list2leda script:
./list2leda edgelist.txt >> graph.gw
Self-loops, edge directions and duplicate edges are ignored.
Figures
The largest common connected component of the best alignment for each pair:
![]() Yeast - Human |
![]() C. jejuni - E. coli |
![]() M. loti - Synechocystis sp. PCC6803 |
![]() F. tularensis - H. sapiens |
![]() B. anthracis - H. sapiens |
![]() Y. pestis - H. sapiens |
Other network alignment algorithms
Notes
All p-values in the paper were calculated with MATLAB 2007. Newer MATLAB versions use a different output precision, so their p-values may differ from those obtained with the 2007 version (e.g. old version: 1.7E-9, newer version: 0.0).





