Quantitative microarray spot profile optimization: A systematic evaluation of buffer/slide combinations

D P Kreil1, R P Auburn, L A Meadows, S Russell, G Micklem
1ISMB03@Kreil.Org, University of Cambridge

The non-trivial choice of a good combination of microarray spotting buffer and substrate slide chemistry is critical for the reliability of microarray hybridization experiments. This choice determines the morphology of spots and, moreover, the variance of that morphology. Both strongly affect how well the extracted intensity signals correspond to sample RNA abundances. Comparisons of different buffer/slide combinations, however, have tended to be subjective, and are not suitable for a systematic approach. Spot morphology and its variance were assessed objectively by measuring the average and variance of radial spot pixel intensity profiles. The variances assessed comprised deviations from radial symmetry and spot-to-spot reproducibility, as well as general pin-to-pin differences and slide effects. A quantitative and systematic evaluation of over 24x6 buffer/slide combinations yielded a new protocol giving superior spot profiles with minimal variance whilst also scoring excellently on other technical considerations, like strong signal intensity, low carry-over, low evaporation rate,and lack of rush. Clearly, this is an optimization problem in a very high dimensional space, and results strongly depend on many other parameters, like probe concentration or probe length distribution. The process introduced here, however, can generally be applied to explore this space and evaluate new materials and technologies as they emerge. The present study is proof of concept work. Specific results for 80ng/Ál cDNA, and for oligo probes will be published elsewhere (submitted; please request a preprint). Online Supplement