Kosow [10] has extended the consistency concepts to concern also consistency within a scenario sample (or scenario set), consistency between different forms of one scenario and consistency of underlying models of scenarios.
In this paper, we investigate redundant probe set consistency in a data set harvested from the Gene Expression Omnibus [9].
I decided to use only the data provided by the DREAM organizers, because of concerns of inter-data set consistency.
A genome-based filter that screens and groups probe sets according to their overlapping genomic alignments significantly improved redundant probe set consistency.
To test the effects of P/A filtering on redundant probe set consistency, we removed probe sets that were deemed as absent by MAS5.0 present/absent call [21] and only analyzed the redundant probe sets that were deemed present.
As a further measure of redundant probe set consistency, we computed Pearson's correlation coefficient between the fold-changes obtained for the two different probe sets belonging to each redundant probe set pair.
However, I decided not to pursue gene regulatory networks for this problem, for reasons stated above regarding to inter-data set consistency, and also because most network reconstruction algorithms are model-driven, relying on simplifying model assumptions that are often hard to be tested or fulfilled.
To further examine the relationship among P/A calls, significant probe sets, and consistency of the redundant probe sets, we examined in detail the genes with two probe sets.
In essence, the three metabolite data sets showed consistency among the data derived from the independent gradients and supported, visually (Figure 2) and statistically (see above; Figure S1), the separation of compartmental-enriched fractions.
The heat map of proteomic alterations from the immunodepleted early and late OA sets show consistency across biological replicates.
