A common assumption in microarray analysis is that measurements such as y jiM and y jiU are lognormally distributed; consequently, is normally distributed, and we expect the technical error introduced by the microarray to occur on the logit scale.
As microarray measurements are noisy, reproducibility of such measurements deserves a careful attention.
We will show later that this natural continuous extension exists as a discretization limit and allows us to handle continuous data such as microarray measurements directly.
The latter two approaches either directly or indirectly assume that the network in question is a static one, and samples of nodal states, such as microarray measurements of gene expressions are i.i.d.i.d
It is remarkable that in the decade or two since their creation, high-throughput molecular measurements, such as microarrays, have already been used to study so many human diseases, and that data from these experiments are publicly available.
To date, the main sources of such data are microarray measurements of genome-wide expression profiles, with over 400,000 profiles stored in GEO [1] alone as of April 2010.
Instead, the relative performance of preprocessing algorithms can be assessed using independent measurements, such as a second microarray platform or quantitative real-time PCR (RT-PCR) [9], [10].
However, given the increasing number of microarray measurements, it may be possible to reconstruct such pathways and uncover missing connections directly from experimental data.
High throughput methods, such as high density oligonucleotide microarray measurements of mRNA levels, are popular and critical to genome scale analysis and systems biology.
Although it may be argued that microarray measurements are merely alternative ways to monitor well-known processes such as proliferation, ER, or HER2 signaling, their results are not perfectly concordant with conventional variables.
