Hybrid Fuzzy Algorithm for Protein Clustering Using Secondary Structure Parameters

Clustering of proteins is important in the field of bioinformatics. Clustering of proteins is used for analyzing the proteins to determine their functions and structure. The number of partitioning techniques, hierarchical methods and graph-based methods are available for clustering protein sequences. In this paper, we propose a hybrid fuzzy technique for clustering proteins based on its secondary structure elements. The algorithm works in two stages. In the first stage, initial number of clusters is determined using k-nearest neighbor distances. The second stage comprises membership calculation and cluster construction. The performance of the hybrid fuzzy clustering was evaluated by comparison with other existing methods on four data sets. Experimental results show that proposed approach performs better in terms of validity indices and execution time as well.