Improved sequence‐based prediction of protein secondary structures by combining vacuum‐ultraviolet circular dichroism spectroscopy with neural network

Abstract

Synchrotron-radiation vacuum-ultraviolet circular dichroism (VUVCD) spectroscopy can significantly improve the predictive accuracy of the contents and segment numbers of protein secondary structures by extending the short-wavelength limit of the spectra. In the present study, we combined VUVCD spectra down to 160 nm with neural-network (NN) method to improve the sequence-based prediction of protein secondary structures. The secondary structures of 30 target proteins (test set) were assigned into alpha-helices, beta-strands, and others by the DSSP program based on their X-ray crystal structures. Combining the alpha-helix and beta-strand contents estimated from the VUVCD spectra of the target proteins improved the overall sequence-based predictive accuracy Q(3) for three secondary-structure components from 59.5 to 60.7%. Incorporating the position-specific scoring matrix in the NN method improved the predictive accuracy from 70.9 to 72.1% when combining the secondary-structure contents, to 72.5% when combining the numbers of segments, and finally to 74.9% when filtering the VUVCD data. Improvement in the sequence-based prediction of secondary structures was also apparent in two other indices of the overall performance: the correlation coefficient (C) and the segment overlap value (SOV). These results suggest that VUVCD data could enhance the predictive accuracy to over 80% when combined with the currently best sequence-prediction algorithms, greatly expanding the applicability of VUVCD spectroscopy to protein structural biology.

Spectroscopic Characterization of Intermolecular Interaction of Amyloid β Promoted on GM1 Micelles

“…In this analysis, the 3 10 -helix was classified as an unordered structure. The root-mean-square deviation ( δ ) and the Pearson correlation coefficient ( r ) between the X-ray and VUVCD estimates of the secondary structure contents of the reference proteins were 0.058 and 0.85, respectively, confirming the high accuracy of the VUVCD estimation [27]. …”

Structural analysis of lysine‐4 methylated histone H3 proteins using synchrotron radiation circular dichroism spectroscopy

“…The computational protocol is described elsewhere. 16 Briefly, we used an NN algorithm that predicts the positions of secondary structures by using evolutionary sequence information based on the position-specific scoring matrices generated by the PSI-BLAST algorithm, 17 referring to the numbers of α-helix and β-strand segments and those of amino acid residues forming α-helix and β-strand structures determined by CD spectroscopy and SELCON3 analyses. Turn and unordered structures estimated by the SELCON3 analysis were classified as coil structures.…”

“…It is known that the accuracy of the VUVCD-NN combination method is about 75% for 30 reference proteins. 16 3 | RESULTS AND DISCUSSION 3.1 | CD spectra Figure 1 shows the CD spectra of K4-methylated H3 proteins. The CD spectrum of H3K4me0 reproduced from the literature data 9 is also shown.…”

Protein characterisation by synchrotron radiation circular dichroism spectroscopy

“…Combinations of SRCD measurements and sequence-based neural network modelling methods have been used to improve the empirical predictions of protein secondary structures (Lees & Janes, 2008 ;Matsuo et al 2008). By providing accurate information on the net amounts of each type of secondary structures as calculated from SRCD spectra to augment the training of a standard neural network for the position-specific prediction of the secondary structure type of each amino acid in a protein, Matsuo et al (2008) found a significant improvement in the prediction of the individual amino acids' secondary structures, when compared the known crystal structure. A similar approach was taken by Lees & Janes (2008), except in the latter case experimental information from both SRCD and FTIR measurements were used to enhance the predictions.…”