Exploring Highly Structure Similar Protein Sequence Motifs using Granular Computing Model based on Adaptive FCM

Protein sequence motifs are very important to the analysis of biologically significant conserved regions to determine the conformation, function and activities of the proteins. These sequence motifs are identified from protein sequence segments generated from large number of protein sequences. All generated sequence segments may not yield potential motif patterns. In this paper, short recurring segments of proteins are explored by utilizing a granular computing strategy. Initially, Fuzzy C-Means (FCM) and Adaptive Fuzzy C-Means clustering algorithms (AFCM) are used to separate the whole dataset into several smaller informational granules and then succeeded by KMeans and Rough K-Means clustering algorithms on each granule to obtain the final results. By comparing the results of two different granular techniques shows that Adaptive FCM granular with Rough K-Means clustering is capable to capture better motif patterns suggests that our granular computing model which combined AFCM granular with Rough K-Means have a high chance to be applied in some other bioinformatics research fields.