Data analysis flow
The general analysis flow for QuantiGene® Plex 2.0 data encompasses three main steps which should be performed on a per-gene basis: background correction of the gene signal, normalization of the gene signal and calculation of fold change in expression (Affymetrix, manufacturer’s manual). In more detail: wells without loaded RNA sample can be summarized in an average background signal (recommendation of three background control wells). Background corrected gene signals are obtained by subtracting the average background signal from the raw median fluorescence intensity (MFI) values. Subsequently, dividing the background corrected values by a normalization gene signal (background subtracted) results in normalized gene signal values. Alternatively, if a set of normalization genes is used, one should divide by the geometric mean of the normalization gene signals. These normalized values are expression ratios (target/housekeeping gene) and will be further referred to as relative expression values. Finally, fold changes in gene expression versus untreated sample are calculated by dividing the relative expression value by the relative expression value of untreated sample (negative control).
Downstream analyses such as dose response modelling rely on qualitative fold change values for proper conclusions. Given the analysis flow as described above, this can only be warranted if accurate, high-quality signals are used throughout all the different data transformations steps (background subtraction, normalization and fold change calculation). QGprofiler includes clear guidelines at each step of the analysis flow in order to ensure maximal quality of the data at each transformation step. The selection of suitable normalization genes and the stability of their corresponding signal is of the utmost importance in this regard.
Quality guidelines