Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: Prediction of protein continuum secondary structure with probabilistic models based on NMR solved structures

Figure 2

KL divergence as a function of test dataset residue entropy. The dashed red and solid blue lines show the KL divergence of predictions on the test dataset (set-286) made by the CCNN and CPNN models, respectively. Residues are binned by secondary structure entropy, and the mean KL divergence of residues in a bin is plotted at the midpoint of the bin. Error bars show plus and minus one standard error around each mean. The numbers of residues in the bins for the 3-class problem are (in order of increasing entropy) 26357, 948, 839, 525, 14, 1036, 489, 4 and 2. For the 8-class problem bin occupancies are 25657, 1878, 777, 1728, 127, 43, 4 and 0.

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us