Referee 1's review:

**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication the manuscript has been revised in light of these comments and to address other editorial requirements.
**********

The authors studied the effect of different the T7-spacer sequences on microarray studies using amplified RNA as a probe. Their data demonstrate that there is a difference in expression levels measured for individual genes caused by homology to the spacer sequence. However, as the authors already state in their discussion, this problem is less pronounced in the newer platforms.

One of the main points that can be raised is how relevant this bias is, since most if not all, researchers focus on analysis of differentially expressed genes between two different samples and thereby they eliminate the experimental bias introduced by the T7 spacer. It is generally accepted that it is not possible to compare results obtained from different platforms. Moreover, most researchers are interested only in the relative ratio compared to a control and not to the absolute expression level, which might be influenced by the homology to the spacer sequences.

Would it be possible to test the effect of the T7 spacer on detection of differentially expressed genes in one or two of the data sets that were used? This could provide further support for the relevance of the T7 spacer bias.

The authors should include the percentages of probes that are effected by the T7 spacer in the different platforms to more accurately demonstrate the impact of this bias.

Would it be possible to design and include a most optimal T7 spacer sequence that would give a low bias in microarray studies?

What is the extra value of figure 4?