Clustering on structured proteins with filtering instances on Bioweka

This paper presents a synthesis on the analysis of structured proteins through data mining techniques. The protein structure is predicted by using target sequences from BLAST. The predicted protein structure of macropus rufus and structured proteolyzed lysozyme protein are interacted together using mergesets filter for clustering. The selection of a clustering technique for protein clustering is a problem. In order to estimate, performance on three clustering algorithms that are SimpleKMeans, ExpectationMaximization and MakeDensityBasedClusterer are analysed. Carried the simulation experiment on open source data mining tool Bioweka. Comparative analysis with various clustering algorithms illustrates the efficiency of the better clustering algorithm. This method can be applied for the large class of applications such as drug discovery, protein-protein analysis, docking, etc