Protein sequence clustering is a process that aims to identify sets of homologous proteins in a protein database. In this paper, two efficient soft c-mediods clustering algorithms for prototype selection for protein sequences are presented. In the proposed techniques patterns are considered to belong to some but not necessarily all clusters. The proposed algorithms is comprised of a judicious integration of the principles of fuzzy sets, semi-fuzzy or soft clustering models, the amino acid mutation matrix. Applying randomized search along with soft clustering model to the fuzzy c-medoids algorithm enables efficient and effective selection of the minimum set of the most informative bio-bases. The efficiency and the effectiveness of the proposed algorithms, along with a comparison with other algorithms, have been demonstrated on different types of protein data sets.

Ease of interpretation of results makes hierarchical clustering algorithms suitable for many applications. However they suffer high computational complexity. Single Linkage hierarchical clustering with its related graph theoretical terms is the most famous among hierarchical algorithms. Several existing techniques reduce its complexity by formulating it as a minimum spanning tree (MST) problem. The main contribution of this research study is minimizing the number of edges given as input to the MST algorithm by first partitioning the dataset using any scalable clustering technique then building k-nearest neighbors table for each partition. After that, outliers and border objects are identified in each partition and moved to a separate partition. Objects in the outliers' partition are used in updating the k-nearest neighbors table of itself and the other partitions. The final edges in the k-nearest neighbors' tables of all partitions are sorted in descending order then merged. The final sorted edges are given as input to a MST algorithm. Several experiments are carried in order to validate the idea, fine tune the input parameters and to get insights of its performance. The proposed algorithm can be completely implemented in parallel which can improve the performance by significant factor.

Biclustering is a key step in analyzing gene expression data by identifying patterns where subset of genes are co-related based on a subset of conditions. This paper proposes a new distance based possibilistic biclustering algorithm (DPBC), in which the average distances between rows and between columns of the bicluster are minimized and at the same time the size of the bicluster is maximized by computing the zeros of the derivative of appropriate objective function. The proposed algorithm uses the possibilistic clustering paradigm similar to another existing possibilistic biclustering algorithm PBC. Whereas PBC is based on residue our approach is applicable to any accepted definition for distances between pairs of rows or columns. Experimental study on the human dataset and several artificial datasets having different noise levels shows that the DPBC algorithm can offer substantial improvements over the previously proposed algorithms.