Combining sequence-based prediction methods and circular dichroism and infrared spectroscopic data to improve protein secondary structure determinations

Table 1 GSEQ dataset 3-fold cross-validation.

Performance measure	α-helix (H)	β-Sheet (E)	Other (G, I, T, B, S, C)
Q2 (%)	89.7	89.2	81.7
Corr	0.770	0.676	0.631
r	0.939	0.901	0.791
abs	0.047	0.047	0.059
δ	0.073	0.069	0.083

Performance parameters from the 3-fold cross-validation of the GSEQ dataset using the neural network sequence-based prediction.

ISSN: 1471-2105