CLIQUE BASED APPROACH TO PREDICT COMPLEXES FROM PROTEIN INTERACTION NETWORK

The ample availability and importance of large-scale protein-protein interaction (PPI) data demand a flurry of research efforts to understand cells' organization, processes, and functioning by analyzing these  data  at  the  network  level.  In  the  bioinformatics  and  data  mining  fields,  network  clustering requires  a  lot  of  attraction  to  discover  clusters  of  interacting  proteins.  Clustering  proteins  in  a  PPI network  has  been  an  excellent  method  for  discovering  functional  modules,  disclosing  functions  of unknown  proteins,  and  other  tasks  in  numerous  research  over  the  last  decade.  In  this  research,  a unique  graph  mining  approach  is  proposed  to  detect  dense  neighborhoods  (highly  connected regions)  in  an  interaction  graph,  including  protein  complexes.  Our  technique  first  finds  size-3 cliques  and  then  expands  these  size-3  cliques  based  on  their  affinity  to  produce  maximal  dense regions. To highlight the efficiency of our suggested strategy, we present experimental results using yeast  and  human  protein  interaction  data.  Our  predicted  complexes  match or  overlap  much  better with the gold standard protein complexes in the CYC-2008 and CORUM benchmark databases than other existing approaches.