Systematic spatial bias in DNA microarray hybridization is caused by probe spot position-dependent variability in lateral diffusion

Abstract

Background: The hybridization of nucleic acid targets with surface-immobilized probes is a widely used assay for the parallel detection of multiple targets in medical and biological research. Despite its widespread application, DNA microarray technology still suffers from several biases and lack of reproducibility, stemming in part from an incomplete understanding of the processes governing surface hybridization. In particular, non-random spatial variations within individual microarray hybridizations are often observed, but the mechanisms underpinning this positional bias remain incompletely explained.

Methodology/principal findings: This study identifies and rationalizes a systematic spatial bias in the intensity of surface hybridization, characterized by markedly increased signal intensity of spots located at the boundaries of the spotted areas of the microarray slide. Combining observations from a simplified single-probe block array format with predictions from a mathematical model, the mechanism responsible for this bias is found to be a position-dependent variation in lateral diffusion of target molecules. Numerical simulations reveal a strong influence of microarray well geometry on the spatial bias.

Conclusions: Reciprocal adjustment of the size of the microarray hybridization chamber to the area of surface-bound probes is a simple and effective measure to minimize or eliminate the diffusion-based bias, resulting in increased uniformity and accuracy of quantitative DNA microarray hybridization.

Figure 1. Schematic layout of the simplified, single-probe microarray.

Hybridization chambers have a square footprint with all sides having a half-length L = 10.8 mm (21.6×21.6 mm² total surface area) and a height H = 0.25 mm. Spot positions (1 to 15) are indicated by the alternating white and grey shading. Spots within each spot position are treated as replicate spots, with decreasing numbers of replicates (n) from spot position 1 (n = 116) to spot position 15 (n = 4). Definitions of diagonal transects (n = 4) and center-line transects (n = 8) are shown as dotted lines.

Figure 2. Spatial variations in fluorescence intensity of labeled genomic DNA hybridized to a genome array of Protochlamydia amoebophila.

Signal intensities from 32 blocks are displayed, with each block composed of 144 (12×12) spots. Signal intensities (expressed in arbitrary fluorescence units) are shown as height and color heatmap in a three dimensional surface plot and displayed as (A) lateral view and (B) top view. An increased boundary signal results in a U-shaped intensity profile across the entire array (A). A tendency for increases in boundary signals is also evident at the boundary of blocks (B), though the pattern is somewhat obscured by additional signal intensity variations caused by different probes at different positions.

Figure 3. Two examples of spatial variation in signal intensities from publicly available microarray images.

Signal intensities are shown as height as well as color heatmap on a three dimensional surface plot. (A) Channel 2 image from a Vibrio cholerae comparative genome hybridization experiment (genomic DNA targets) on a microarray consisting of 16 (4×4) blocks and 272 (17×16) spots per block (ExpID 68809) (B) Channel 1 image from a comparative gene expression hybridization experiment (cDNA targets) on a Mycobacterium tuberculosis microarray that consists of 16 (4×4) blocks and 289 (17×17) spots per block (ExpID 75165). Both arrays were hybridized overnight without agitation. Complete experimental details are available at the Stanford Microarray database under the associated experiment ID.

Figure 4. Probe spot position-dependent spatial variation in signal intensity in the simplified microarray hybridized with labeled RNA (single-color) under target limiting conditions for 18 h.

(A) Mean signal intensities of three replicate hybridizations using Cy3 labeled RNA are shown as height in a three dimensional surface plot. (B) Microarray hybridization of Cy3 (red) and Cy5 (blue) labeled target RNA. Mean relative signal intensities are given as fraction of the spot position with highest absolute signal intensity and are displayed as function of the respective position. Error bars represent relative standard deviation per spot position for three replicate hybridizations.

Figure 5. Comparison of relative signal intensities of diagonal (black) and center line (white) transects on the simplified microarray (see Fig. 1).

Labeled RNA was hybridized under target limiting conditions for 18 h. Mean relative signal intensities for each position are normalized to the lowest value of all positions. Error bars represent relative standard deviations of the mean signal intensities per spot position for three replicate hybridizations.

Figure 6. Factors affecting the boundary bias and implications for signal ratios in competitive hybridizations.

(A) Hybridization of labeled RNA (single-color) at a range of concentrations (75–4,500 ng) for 18 h. (B) Hybridization of labeled RNA (single-color) under target limiting conditions for 18, 65 and 140 h. (C) Competitive two-color hybridization of Cy3- and Cy5 labeled RNA mixed at approximately 3∶1 ratio (900 ng and 300 ng, 18 h hybridization). (D) Ratios of signal intensities from competitive hybridization in C. Signal intensities for A and C are normalized to the spot position with the highest signal and error bars for A to D indicate standard deviation per spot position for three replicate hybridizations.

Figure 7. Numerical simulation of the three-dimensional unsteady diffusion equation to simulate flux of labeled RNA to each probe position over a 24 h hybridization.

Three different cases were considered: a large and shallow chamber, similar in dimensions to the simplified array used in experiments (‘default configuration’, black), a large and deep chamber (‘deep configuration’, red) that tests the effect of increasing the chamber height, and a small and shallow chamber (‘small configuration’, green) that tests the effect of unspotted surface area. The number of spots and the spotted area are the same for all simulations.

Figure 8. Effect of eliminating unspotted surface area on the probe spot position-dependent signal intensity.

300 ng of Cy3-labeled RNA was hybridized for 18 h. The hybridization chamber geometry was the same as previous experiments. The spot grid of the simplified 30×30 array was expanded to reach the edges of the hybridization chamber, which yielded an array of 5,040 spots (72×70). Spot position was calculated identically as for the 30×30 array, with the outer position being one and the inner position equaling 35. Signal intensities are normalized to spot position one and error bars indicate standard deviation per spot position for five replicate hybridizations.