FUZZY C-MEANS CLUSTERING FOR SNP HAPLOTYPE RECONSTRUCTION PROBLEM

In this paper, we proposed an algorithm which uses fuzzy clustering method in order to solve haplotype assembly problem. Haplotypes contain essential information which is useful for disease association studies. However, determining haplotypes by experimental methods is an expensive and time-consuming task. The proposed method constructs a graph based fuzzy conflict on the similarities of all input fragments and next, the cluster centers are used as initial centers by fuzzy c-means (FCM) algorithm. Experimental results show that our method outperforms and achieves a suitable accuracy in most of the cases.