Clustering Time-Series Gene Expression Data with Unequal Time Intervals

摘要: Clustering gene expression data given in terms of time-series is a challenging problem that imposes its own particular constraints, namely exchanging two or more time points not possible as it would deliver quite different results, and also lead to erroneous biological conclusions. We have focused on issues related clustering temporal profiles, devised novel algorithm for profile microarray data. The proposed method introduces the concept alignment which achieved by minimizing area between aligned profiles. overall pattern context accomplished applying agglomerative combined with alignment, finding optimal number clusters means variant index, can effectively decide upon dataset. effectiveness approach demonstrated well-known datasets, yeast serum, corroborated set pre-clustered genes, show very high classification accuracy method, though an unsupervised scheme.