Non-swarm-based computational approach for mining cancer drug target modules in protein interaction network

Cancer is a lethal disease that drew the entire world over the past decades. Currently, numerous researches focused on these cancer treatments. Most familiar among them is the targeted therapy; a customized treatment type depends on the cancer drug targets. Further, the selection of targets is a quite sensitive task. The computational approaches are lagging in this field. This paper is intended to propose an optimized multi-functional score–based co-clustering with MapReduce (MR-CoC opt ) approach for drug target module mining with optimal functional score set selection. This approach uses biological functional measures for co-clustering, MapReduce framework for handling redundant modules and complex protein interaction network (PIN), and non-swarm intelligence algorithm-bladderworts suction for optimal functional score set selection. It extracts the cancer-specific drug target modules in protein interaction networks. The protein complex coverage of the results is compared with the existing approach. The biological significance of the results is analyzed for the presence of cancer drug targets and drug target characteristics. From these results, novel cancer drug target modules are suggested for the targeted therapy and the active pharmaceutical drugs for these modules are also highlighted. Graphical abstract