Clustering Temporal Gene Expression Data with Unequal Time Intervals

We have focused on the problem of clustering time-series gene expression data. We present a novel algorithm for clustering gene temporal expression profile microarray data, which is fairly simple but powerful enough to find an efficient distribution of genes over clusters. Using a variant of a clustering index can effectively decide upon the optimal number of clusters for a given dataset. The clustering method is based on a profile-alignment approach, which we propose and that minimizes the (square) area between two aligned vector profiles, to hierarchically cluster microarray time series data. The effectiveness of the proposed approach is demonstrated on two well-known, yeast and serum.