One of the major challenges for microarray analysis is the design of probes specific to a given target. This is largely due to the conserved nature of many genes and the large amount of unknown sequence data present in many samples. Although longer probes may increase sensitivity, they also increase the potential for cross-hybridization with nontarget sequences. By using oligo probes, it is possible to avoid conserved regions of genes or areas containing stable secondary structure during the probe design process. The shorter oligo probes (~20-mers) can differentiate a single mismatch in a probe-target hybridization, making them ideal for use with highly conserved genes such as 16S rRNA in POAs (Wilson et al., 2002a; Urakawa et al., 2003; Zhou et al., 2004). A common format for these arrays includes sets of probes that perfectly match a target sequence and corresponding sets of probes containing one mismatched nucleotide, usually at a central position. Greater signal intensity for the perfect probes versus the mismatched probes indicates detection of the target sequence. Even though the mismatched probes typically have greatly decreased ability to bind the target of interest (Zhou et al., 2004), spurious results are sometimes obtained. This is likely due to the presence of similar yet unknown sequences and can make it difficult to achieve complete discrimination. One way to address this problem is to design and use multiple perfectly matched and mismatched probe combinations for each organism or gene of interest. The results from the probe pairs are then compared statistically, and those with abnormal results (higher signal intensity for the mismatched probe) are discarded during data analysis. It may also be possible to improve the differentiation of perfectly matched and mismatched probes by determining the thermal dissociation curve for each probe-target hybridization on a three-dimensional array platform (Liu et al., 2001; El Fantroussi et al., 2003; Urakawa et al., 2003), but this may be difficult for high-density planar arrays given the current technology.
Most functional genes are more variable than rRNA genes thus enabling the use of longer oligo probes (~40- to 70-mers), which have greater sensitivity, while still achieving species-level specificity. These longer oligo-based probes have been reported to discriminate sequences less than 80-90% similar to the probes (Taroncher-Oldenburg et al., 2003; Rhee et al., 2004). Specificity can also be increased or decreased, to an extent, by adjusting the stringency of the hybridization conditions (temperature, formamide concentration, salt concentration, etc.) (Wu et al., 2004). However, caution should be exercised when using an array under more or less stringent conditions than that for which it was designed, as this could cause overestimated or underestimated results and ultimately inaccurate conclusions.
Was this article helpful?