Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2008;3(10):e3400.
doi: 10.1371/journal.pone.0003400. Epub 2008 Oct 15.

ANGLOR: a composite machine-learning algorithm for protein backbone torsion angle prediction

Sitao Wu  1 , Yang Zhang
Affiliations

ANGLOR: a composite machine-learning algorithm for protein backbone torsion angle prediction

Sitao Wu et al. PLoS One. 2008.

Abstract

We developed a composite machine-learning based algorithm, called ANGLOR, to predict real-value protein backbone torsion angles from amino acid sequences. The input features of ANGLOR include sequence profiles, predicted secondary structure and solvent accessibility. In a large-scale benchmarking test, the mean absolute error (MAE) of the phi/psi prediction is 28 degrees/46 degrees , which is approximately 10% lower than that generated by software in literature. The prediction is statistically different from a random predictor (or a purely secondary-structure-based predictor) with p-value <1.0 x 10(-300) (or <1.0 x 10(-148)) by Wilcoxon signed rank test. For some residues (ILE, LEU, PRO and VAL) and especially the residues in helix and buried regions, the MAE of phi angles is much smaller (10-20 degrees ) than that in other environments. Thus, although the average accuracy of the ANGLOR prediction is still low, the portion of the accurately predicted dihedral angles may be useful in assisting protein fold recognition and ab initio 3D structure modeling.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. Flowchart of ANGLOR for the phi and psi angle predictions.
Three sets of features, position-specific scoring matrix (PSSM), secondary structure (SS) and solvent accessibility (SA), are used as inputs of two machine-learning predictors (neural networks and support vector machines) for phi and psi separately.
Figure 2
Figure 2. Two simplified models with the output Y generated from random processes for a given input X in [0, 30].
(A) Training data generated from the random fluctuations around four horizontal line segments; (B) training data generated from the random fluctuations around two sine waves of the frequency equal to 1/2π and 1/π respectively; (C) testing data (solid) and prediction results by two training predictors of SVM prediction (dashed) and NN prediction (dotted) for the model from A; (D) testing data (solid) and prediction results by two training predictors of SVM prediction (dashed) and NN prediction (dotted) for the model from B; (E) histogram of Y from A; (F) histogram of Y from B.
Figure 3
Figure 3. Ramachandran plot and histograms of phi and psi angles calculated from residues in 500 non-homologous training proteins.
(A) Ramachandran plot; (B) histogram of phi angles; (C) histogram of psi angles. Alpha-helix, beta-strand and polyproline-II are represented by “α”, “β” and “P” respectively.
Figure 4
Figure 4. The comparison of predicted (dotted lines) and experimental values (solid lines) of phi and psi angles for three typical alpha-, beta-, and alpha/beta-proteins.
Secondary structures of the proteins are signified at the lower part of each box, with coil, beta-strand, and alpha-helix residues represented by thin lines, thick lines, and thick curves, respectively. (A) phi angle for 1n7sD; (B) psi angle for 1n7sD; (C) phi angle for 1k5nB; (D) psi angle for 1k5nB; (E) phi angle for 1lj9B; (F) psi angle for 1lj9B.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Branden C, Tooze J. 1999. Introduction to protein structure: Garland Publishing, Inc.
    1. Neal S, Berjanskii M, Zhang H, Wishart DS. Accurate prediction of protein torsion angles using chemical shifts and sequence homology. Magn Reson Chem 44 Spec No. 2006:S158–167. - PubMed
    1. Berjanskii MV, Neal S, Wishart DS. PREDITOR: a web server for predicting protein torsion angle restraints. Nucleic Acids Res. 2006;34:W63–69. - PMC - PubMed
    1. Wood MJ, Hirst JD. Protein secondary structure prediction with dihedral angles. Proteins. 2005;59:476–481. - PubMed
    1. Mooney C, Vullo A, Pollastri G. Protein structural motif prediction in multidimensional phi-psi space leads to improved secondary structure prediction. J Comput Biol. 2006;13:1489–1502. - PubMed

Publication types