CAG: A java application to identify protein complexes in PPI networks based on core-attachment approach incorporating Gene Expression Profile

[The paper of CAG is yet to be submitted for publication.]

Author: Seketoulie Keretsu.

Description: CAG is based on a core-attachment approach to detect protien complexes with high density and high correlation core proteins. 
The cores are also supported by attachment proteins with high connectivity with the core proteins.                                           
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
\\\\\\\\\\\\\\\\\\INPUT DATA\\\\\\\\\\\\\\\\\\\\\\
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
PPI input: Weighted PPI network. Where interaction are given as YAL001C	YBR123C	0.983749. 
Where YAL001C and YBR123C is the systematic name of the proteins seperated by tab or space. 
The weight is give after the two protein names with a tab seperating it.[collins2007]

GEP input: The Gene Expression Value of the Proteins with time course.[gene.txt]

//////////////////Parameters:////////////////////////
Density: The threshold value of density to consider a cluster as a core.
Similarity: The minimum threshold similarity value to add a protein to a cluster to form a complex core.


\\\\\\\\\\\\\\\Input Data Example///////////////////
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
PPI input: collins2007.txt
GEP input: gene.txt
Benchmark input: sgd.txt
Density: 0.6
similarity: 0.1 [to purge only the negative and low similarity values.]
 
//////////////OUTPUT///////////////////////////////
output on interface:

Precission:
Recall:
F-measure:
Detected: number of Predicted complexes that matched with a real complex in the reference benchmark data.

Output File: 
The complexes predicted that matched with a real comples in the reference data given in [output_DetectedComplex.txt]


NOTE: The CAG method interface is expected to be submitted along with the CAG paper for publication after adding some additional design components.