Polynomial equation models for yeast cell-cycle time series microarray data by analysing fidelity matrices of gene expression values

Abstract

Identifying the various gene expression response patterns in microarray time-course experiments is a very challenging issue. It is impossible to manually characterise a parametric form for each of the time point values in gene microarray in a gene by gene manner. The difficulty arises due to heterogeneity in the regulatory reactions among thousands of genes. In this paper, we develop two polynomial equation models to automatically define the time-dependent expression patterns for gene expression time series microarray data. Our developed polynomial equations with power terms from a set of fixed values offer a wide range of curve shapes. The generated polynomial equations suggest the best fitting model of gene microarray time series data values. It also measures the level of mRNA expression of thousands of genes. The effectiveness of the proposed models is first demonstrated for artificial data sets. We then identify the best suited polynomial equation of our specimen real-life gene microarray time series data of yeast cell cycle.