outputAlignment = MONACO (netIdList, inputFolder, idFlag, outFileName, alpha, alignmentConstructionStrategy)
Input netIdList ID list of input networks.
inputFolder Input folder.
idFlag This flag is used to expedite the reading process of input files.
id_flag = 1, if every node id is represented as "networkID+number".
id_flag = 0, otherwise.
outFileName Output file name including path.
alpha A balancing parameter between node-level similarity and topological similarity.
alpha < 0.5 works well in most cases.
alignmentConstructionStrategy 0: Many-to-Many mapping
1: One-to-one mapping
2: Maximum Weighted Bipartite Matching (Pairwise network alignment only)
output outputAlignment Alignment results
Suppose we have three PPI networks 'a', 'b', 'c'.
To run MONACO, the files listed below are required.
Tab-separated undirected PPI network files: a.net, b.net, and c.net
Example) a.net
a1 a2
a3 a1
a4 a2
a2 a3
Node-level similarity score file for each network pair: a-b.sim, a-c.sim, b-c.sim
Example) a-b.sim
a1 b1 153
a1 b3 55
a1 b7 49
a2 b3 444
a3 b3 211
a3 b4 122
a4 b5 251
a4 b8 71
* Nodes in the first column must be the nodes of the first network in the file name.
Similarly, nodes in the second column have to be the nodes in the second network of the file name.
Each line of the output file corresponds to individual cluster
Example)
a4 b5 c4
a1 b1 b7 c1
a2 b3 c2
a3 b4 c3
alignment = MONACO({'a', 'b', 'c'}, 'test', 1, 'output.txt', 0.4, 1)
For more information on the algorithms, please see:
Hyun-Myung Woo and Byung-Jun Yoon (2019) MONACO: accurate biological network alignment through optimal neighborhood matching between focal nodes
Contact: bjyoon@ece.tamu.edu