justTheCliques: Find maximal cliques, via the Bron Kerbosch algorithm, Bron Kerbosch in Wikipedia
cp5: Fast clique percolation algorithm, described in Cornell University Library
Copyright 2009-2011 - Aaron McDaid [email protected]. Licensed under GPL v3. See gpl.txt included with this package.
First, to get your hands on the latest code from GitHub. In a directory of your choice:
git clone --recursive https://github.com/aaronmcdaid/MaximalCliques.git
make clean justTheCliques cp5
./justTheCliques your_edge_list.txt [--stringIDs] > cliques.txt
./cp5 your_edge_list.txt output_directory [--stringIDs]
- justTheCliques will output the cliques to standard output, hence you should redirect it with "> cliques.txt" as in this example.
- cp5 will create its output in a directory of your choice.
or, if you just want cliques with at least 10 nodes in them,
./justTheCliques your_edge_list.txt -k 10 [--stringIDs] > cliques.txt
this should be a little faster. --stringIDs is to allow strings, not just integers in the input - see below.
The cliques themselves are printed to stdout (hence the redirection above). Various summary stats are printed on stderr. So, if you didn't want the cliques but did want to see the sizes of the cliques found, to
./justTheCliques edge_list.txt > /dev/null
Each line of the your_edge_list.txt represents an edge. The first two fields (delimited by commas, pipes(|), spaces or tabs) are the names of the two nodes that are connected. Directionality is ignored, and self-loops will be rejected. Any other fields on the line are also ignored.
By default, the node names are integers (64-bit integers). But you can change this with the --stringIDs option; note this will increase memory usage.