Armed with the mean and standard deviation of both distributions and knowing the total number of data points on the array one can evaluate the expected p-value for copy number aberrations affecting a given number of probes with the same mathematical formulation that is used to calculate a t-test.
V1=Percentage of data points in linear range (as defined by the interval [ ar, M / r ]) of all data points on the array, where a is the background level, M is the saturation level, r is the threshold value (as described earlier).
Also, since we do not force a percentage of points on the array to be called as outliers, then we can be confident that the iterative process will terminate in reasonable time (although a maximum number of iterations can be specified).
Specifically, we focus on modeling the uncertainty associated with the location of the point source on the array.
At this moment, the weighting function is introduced, which essentially describes the sensitivity of an infinitesimal point located on the array at.
The analysis was performed on three randomly chosen fields for each time point and cell line, on the array made from cell pellets, and an average was calculated.
For each probe, the difference between the intensities on two arrays is plotted on the vertical axis and the average of the two intensities is plotted on the horizontal axis; there is a data point for each probe on the array.
Taking into consideration dye-swaps and technical replicates, each sample was hybridized six times and there were a total of 72 data points for each gene on the array at 1 d.p.i. and 36 at 5 and 15 d.p.i.
Because the natural world consists of extended objects – and not of independent points - the visual images projected on the array of photoreceptors of exploratory animals are redundant [2], [16], dominated by slow intensity changes (low frequency or 1/f-type of correlations: [3], [9], [60]).
Each point on the curve was obtained by averaging the signal from an array with four replicates.
