Two heuristic approaches to describe periodicities in genomic microarrays

Jörg Aßmus, Hans Arnfinn Karlsen, Dag Tjøstheim


  • In the first part we discuss the filtering of panels of time series based on singular value decomposition. The discussion is based on an approach where this filtering is used to normalize microarray data. We point out effects on the periodicity and phases for time series panels. In the second part we investigate time dependent periodic panels with different phases. We align the time series in the panel and discuss the periodogram of the aligned time series with the purpose of describing the periodic structure of the panel. The method is quite powerful assuming known phases in the model, but it deteriorates rapidly for noisy data.


Full Text:



Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.