analysis on microdissected material. One way to deal with this is to amplify the RNA of the sample, either using a homemade RNA amplification method[11] or one of the commercially available kits (i.e., www.agilent.com). The most challenging, yet difficult, part of amplifying an RNA sample is to do it linearly—meaning that the relative amount of all different RNA in the amplified sample is the same as in the original sample. Although many RNA amplification kits claim to perform linear amplification, in reality, when an unamplified sample is compared with its amplified RNA for many genes, a ratio other than 1 is obtained. As long as the nonlinearity is reproducible among all amplified samples that are compared in an array experiment, such an experiment will produce valid results. However, one should realize that all manipulations performed on the RNA samples will introduce extra noise in the results due to the limited stability of the RNA molecule. Moreover, the oligonucleotide libraries offered by various companies have been designed using the sense coding strand of the RNA. Using amplified RNA on these oligonucleotides enforces hybridization with unstable RNA, instead of cDNA. Although methods have been published to amplify even degraded RNA,[12] we would argue to avoid RNA amplification, if at all possible.

Instead, another approach used to deal with small size samples could be optimizing the yield of the cDNA labeling reaction and lowering the threshold for the amount of input RNA needed to achieve reasonable results in a microarray experiment. By optimizing column purification steps and hybridization conditions, we have scaled down the amount of input RNA by more than 10-fold—to as low as 4 mg of total RNA with very limited loss of signal intensity and reproducibility of the data. Whereas many researchers are using cover slips for their hybridization or hand-made hybridization chambers, automated hybridization, which enables moving targets in low volumes and very efficient washing, has been developed. All these measures increase the sensitivity of

Fig. 2 Channel balancing omits the need for Lowess correction. Scatter plots showing the results of two microarray experiments using a human tumor cell line (Cy3 channel) and Human Universal Reference RNA (Stratagene) (Cy5 channel). The first experiment was performed without channel balancing (A) and results were corrected by Lowess smoothing (B). (From Ref. [15].) The second experiment was performed with channel balancing (C) showing that Lowess correction is not necessary. Arrays were performed by manual hybridization on 19 K human oligo arrays (Sigma/Compugen) as described in the legend of Fig. 1. (From Ref. [16].) Spots were measured using ImaGene software (BioDiscovery, USA). All plots have been constructed using signal mean values and all spots including flagged ones are included on the plots. (View this art in color at www.dekker.com.)

Was this article helpful?

0 0
Getting Started With Dumbbells

Getting Started With Dumbbells

The use of dumbbells gives you a much more comprehensive strengthening effect because the workout engages your stabilizer muscles, in addition to the muscle you may be pin-pointing. Without all of the belts and artificial stabilizers of a machine, you also engage your core muscles, which are your body's natural stabilizers.

Get My Free Ebook

Post a comment