Simulation of DNA array hybridization experiments and evaluation of critical parameters during subsequent image and data analysis

Table 1 Definition, modelling and critical effects of simulation parameters.

Parameter	Model	Variation	Critical effect⁽¹⁾
Spot variation	spot shift (Gaussian distribution)	SD from ideal position	SD > 0.15–0.2 mm 16.7–22.2%⁽²⁾
Pin variation	block shift (Gaussian distribution)	SD from ideal position	SD > 0.12–0.167 mm 13.3–18.6 %^(2,3)
Spot shape	a) two-dimensional Gaussian distribution	a) no variation (fixed SD = 0.1482 mm)
	b) Crater spot distribution	b) radius of crater	b) radius > 0.1995 mm 22.2 %(^2,4,5)
	c) Plateau spot distribution	c) no variation (fixed radius of cylindric plateau spot = 0.342 mm)
Global background	additive signal from a Gaussian distribution	fixed mean/SD derived from experimental data	not critical⁽⁶⁾
Local background	additive signal from fractal clouds	signal/background ratio	mean signal/background ratio < 25

⁽¹⁾Pearson correlation < 0.95. ⁽²⁾ Percent of spot radius relative to the mean spot distance. ⁽³⁾ For VisualGrid and FA; AIDA did not become critical for the parameter range used for the simulations in this paper. ⁽⁴⁾Only analysed with FA. ⁽⁵⁾For radius ≥ 0.228 mm the automatic gridfind failed. ⁽⁶⁾Not critical for global background noise that is comparable to our experimental reference data.

ISSN: 1471-2105