Figure 1
From: 'maskBAD' - a package to detect and remove Affymetrix probes with binding affinity differences
Principle of the masking method. Each dot represents a sample— human or chimpanzee. On the axes are fluorescence intensities for the probes from the same probe set. Our method performs a t-test of the slopes to each point (green and blue lines), assuming that the intercept is taken from all points (black line). On the left ( a) there is no binding afinity difference between humans and chimpanzees in either probe. On the right ( b) one of the probes is BAD.