ABSTRACT
In this chapter, the applications of system identication theory and nonlinear factor analysis in the processing of the time series of DNA microarrays are described.*
Analyzing time series of molecular measurements, such as the DNA microarray time series, is a major step in addressing important problems in molecular biology. The gene expression is a temporal process, as the mRNA of each gene is affected by the mRNAs of other genes in previous time steps. Moreover, when a perturbation is applied to almost any biological system, the expression values of the majority of genes will be affected in the future times. Thus, to make an accurate analysis of the functional role(s) played by each gene, it is necessary to analyze time series DNA microarray measurements instead of static data. Unlike static microarray data, time series microarray contains the information regarding the dynamic correlations and interactions among genes that occur in time. However, as mentioned in most of the early works, in analyzing time series data, use the methods that have been designed and specialized for static data.
