The global normalisation factor for a given array in the batch was then calculated as the geometrical mean of all its individual pair-wise normalisation factors with the other arrays in the batch.
In the second step of TileProbe model building, all samples from a given array platform in GEO are grouped according to studies and experimental conditions.
Usually, the synthesis is considered in a given number of arrays and array response, determining the array element position and incentive distribution.
In the range query problem, we specify a range of a rectangular shape in a given (n,n) array, and count the number of points in the range.
A linear model was fitted for each gene in a given series of arrays by using the lmFit function.
We expect the suitable range of parameters to depend on the array platform as well as statistical properties of the array profiles in a given data set.
We pick up in E (P k ) the point P i which is the closest to the point Q i of a given optimal array, in the sense that the angle P i OQ i (O being the sphere center) is the smallest (Fig. 1).
The GC-RMA algorithm is based on a linear additive model, and thus considers all arrays in a given dataset when estimating expression values for each chip, unlike the GCOS algorithm.
According to Affymetrix microarray guidelines, comparable scaling factors between arrays in a given experiment are critical to minimizing differences in overall signal intensities, thereby allowing for more reliable detection of biologically relevant changes [ 35, 52].
The first 250 probe sets were excluded from the regression procedure to ensure that the variation of the mean expression of two arrays in a given consecutive sample is small.
