Extending an algorithm for clustering gene expression time series
Mikko Korpela and Jaakko Hollmen
In: Probabilistic Modeling and Machine Learning in Structural and Systems Biology (PMSB 2006), 17-18 Jun 2006, Tuusula, Finland.
The development of microarray technology has enabled simultaneous expression measurements from tens of thousands of genes. Many gene expression experiments produce time series data with only a few (around 5) time points, due to the high measurement costs. The time series usually represent the dynamic response of an organism to a change in conditions, e.g. application of some drug or other treatment. Here we share some of the experiences we gained while analyzing such data sets, originating from a collaborative project. More information about that work can be found in. We focus on a particular clustering algorithm designed for short time series data. We found that some inaccuracies in the original presentation of the algorithm need to be addressed. In addition to providing corrections for the problems, we also present an extension to the algorithm.