Ian J. Laurenzi and Erdem Arslan. Chemical Engineering, Lehigh University, B330 Iacocca Hall, 111 Research Dr., Bethlehem, PA 18015
Today, scientists use a variety of microarray assays for biological investigations such as expression profiling. However, results of investigations using different platforms may not be the same due to the differences of the assay designs (i.e. oligonucleotide probe sequences). Although the Microarray and Gene Expression Data (MGED) Society recommends that certain information be provided with microarray data to assist in microarray data standardization, implementation of that information is a considerable challenge: quantification of the complex nature of the underlying hybridization reaction network yields either exceptionally coupled population balance equations. In this paper, we evaluate a number of oligonucleotide microarray design features using a computational method that we have developed for simulation of microarray experiments, namely, the number of reporters (probes) per gene, and the inclusion of mismatched (MM) probes for the ostensible development of a cross-hybridization-free differential signal (PM-MM). This method permits the evaluation of the sensitivity and specificity of each probe as well as the specific cross-hybridizing cDNA associated with each. Lastly, we use our simulation technique to evaluate the relative fidelities of probe sets designed using different oligomer design tools. Our results illustrate the utility of the Minimum Information About a Microarray Experiment (MIAME) criteria for engineering and standardization of microarrays and suggest methods for standardization of microarray technology via the robust population and thermodynamic quantification of microarray assays.