Clustering microarray time-series data using expectation maximization and multiple profile alignment

摘要: A common problem in biology is to partition a set of experimental data into clusters such way that the points within same cluster are highly similar while different very different. In this direction, clustering microarray time-series via pairwise alignment piece-wise linear profiles has been recently introduced. We propose EM approach based on multiple natural cubic spline representations gene expression profiles. The achieved by minimizing sum integrated squared errors over time-interval, defined Preliminary experiments well-known 221 pre-clustered Saccharomyces cerevisiae yield encouraging results with 83.26% accuracy.