Figure 3

nSVM's workflow. MudPIT is applied to acquire mass spectrometry data from a biological system in different states (1). The data are subsequently identified by SEQUEST and filtered by DTASelect (2). nSVM is applied to pinpoint differences in the protein expression profiles by using a GA (3.2). Each individual's genome is an array of bits (3.3) that corresponds to a set of proteins (3.1 and 3.2) that will be selected from the dataset (3.4) to be evaluated as a solution (3.5) according to their spectral counts.