Counter-selection facilitated plasmid construction by homologous recombination in Saccharomyces cerevisiae.

Vol. 35, No. 4 (2003) BioTechniques 693 points are distributed along lines that represent the expected differences in signal intensity from each channel. Without a systematic dye effect or after proper normalization, we would see three sets of data points along three equally spaced lines parallel with slope one. On the other hand, the direct comparison of measurements for each dye on different slides can be interpreted with no normalization because our experimental setting does not require across-slide normalization. Therefore, the most direct way to assess the efficiency of RNA amplification and Cy3 and Cy5 dye incorporation during RNA labeling is to directly compare measured quantities for each channel on pairs of slides. Figure 2C illustrates the results of these comparisons. For each channel, we show the estimated slope for the corresponding intensity comparison, which corresponds precisely to the expected values. This procedure can be used to monitor the quality of targets after the very last step of the labeling protocol, without compromising the amount of labeled target available for hybridization. Importantly, other commercially available products can efficiently control for amplification but cannot determine the efficiency of the labeling reaction because this would require laser excitation or equivalent wavelengths. Furthermore, one can determine label efficiency using data (Figure 1C) to correct for the specific activity of labeled targets to facilitate data normalization. In the case of greater differences in labeling efficiency, one can repeat the labeling step prior to hybridization, thus avoiding data of lower quality that could impose potential distortions during normalization and eliminating the costs of hybridization.