Figure 5

Learning curve for peptide retention time prediction. This plot demonstrates the squared correlation coefficient depending on the number of training samples for the union of vds1 and vds2. For every training sample size, we randomly selected the training peptides, and 40 test peptides and repeated this evaluation 100 times. The plot shows the mean squared correlation coefficients of these 100 runs for every training sample size as well as the standard deviation for the POBK and the methods introduced by Klammer et al. [16] using the RBF kernel as well as the models by Petritis et al. [13, 14]. The vertical line corresponds to the minimal number of distinct peptides in one of our verified datasets which was acquired in one run.