According to the Affymetrix GeneChip Expression Analysis Manuel, all GeneChip expression arrays are designed to have a minimum sensitivity of 1:100,000. In other words, this concentration ratio corresponds roughly to a few copies of a transcript per cell, or an approximate 1.5 pM concentration.
For standard eukaryotic expression arrays, it is currently recommended that 10 µg of total RNA should be used. It should be cautious that when the amount of total RNA used in the standard assay is reduced, a subsequent decrease in sensitivity will be observed.
It is impossible to obtain 10 µg of total RNA under certain circumstances, such as total RNA isolated from laser capture microdissection samples or tissues obtained from small organisms. In order to get around the problem of low starting materials, it is necessary to apply a two-cycle or three-cycle cDNA synthesis procedure. If the amount of total RNA is in the range of 1 ng to 100 ng, it is required to perform a two-cycle cDNA synthesis. If the amount of total RNA is in the picogram range, a three-cycle cDNA synthesis will be required.
The recommended quantity of biotin-labeled cRNA for one GeneChip hybridization is 10 µg in 200 µl hybridization buffer. Although there is a tolerance for some variation in quantity, hybridization of significantly less cRNA results in reduced sensitivity, particularly for low-copy transcripts.
In order to perform a meaningful statistics analysis, we highly recommend at least three replicates for all conditions. Although microarray experiments could be labor-intensive and expensive, more replicates grant you higher statistical power when it comes to microarray data analysis. Under some circumstances when cost and labor are concerned, it is necessary to develop a statistical method to overcome the limitation of a low replicate number (2-3). This is to say that the statistical algorithm we use at CyberArrayAnalysis is able to handle low-replicate (2-3) data.
