A systematic spatial bias in microarray hybridizations caused by probe spot position-dependent variability in lateral diffusion

The hybridization of nucleic acid targets with surface-immobilized probes is a widely used assay format for high-throughput detection of many parallel targetsin medical and biological research. Though commonly applied, microarray technology still suffers limitations arising from problems of data robustness and reproducibility across platforms, stemming in part from an incomplete understanding of the complex processes governing surface hybridization behavior. It has been observed that there are non-random spatial variations within individual microarray hybridizations, but the causative mechanisms of positional bias remain largely unexplained. This study identifies a symptomatic spatial bias in surface hybridization signal intensitywith systematically increased signal intensities of spots located at the boundaries of the spotted areas of the microarray slide and characterizes the underlying mechanistic principle of this bias using a simplified block array format. Experimentally-derived hybridization dynamics are compared with a mathematical modeling analysis, which together showthat the driver of the spatial bias is a position-dependent variation in lateral diffusion. Numerical simulations employing a diffusion-based model are used to demonstrate the strong influence of microarray well geometry on this spatial bias and to determine optimal conditions for which this bias can be minimized or eliminated, resulting in increased uniformity of microarray hybridization. A simplified square array of 900 identical spotted probes was hybridized using different target concentrations and different time periods to analyze systematically spatial biases in microarray hybridizations. Whole genome oligonucleotide microarray for Protochlamydia amoebophila. A simplified square array of 5040 identical spotted probes was hybridized using 300 ng of Cy3 labeled RNA to analyze systematically spatial biases in microarray hybridizations.