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
. 2006 Apr;15(4):900-13.
doi: 10.1110/ps.051799606. Epub 2006 Mar 7.

Identification of correct regions in protein models using structural, alignment, and consensus information

Björn Wallner  1 , Arne Elofsson
Affiliations

Identification of correct regions in protein models using structural, alignment, and consensus information

Björn Wallner et al. Protein Sci. 2006 Apr.

Abstract

In this study we present two methods to predict the local quality of a protein model: ProQres and ProQprof. ProQres is based on structural features that can be calculated from a model, while ProQprof uses alignment information and can only be used if the model is created from an alignment. In addition, we also propose a simple approach based on local consensus, Pcons-local. We show that all these methods perform better than state-of-the-art methodologies and that, when applicable, the consensus approach is by far the best approach to predict local structure quality. It was also found that ProQprof performed better than other methods for models based on distant relationships, while ProQres performed best for models based on closer relationship, i.e., a model has to be reasonably good to make a structural evaluation useful. Finally, we show that a combination of ProQprof and ProQres (ProQlocal) performed better than any other nonconsensus method for both high- and low-quality models. Additional information and Web servers are available at: http://www.sbc.su.se/~bjorn/ProQ/.

PubMed Disclaimer

Figures

Figure 1
Figure 1
The distribution of S-score and residue displacement. All residues were grouped by S-score and residue displacement relative to the STRUCTAL structural alignment. Residues off are the number of residues the alignment is shifted (displaced) compared to structural alignment (data taken from the hmtest set).
Figure 2
Figure 2
Finding the optimal sequence window size for ProQres. Performance for neural nets trained using different window size to calculate the structural parameters.
Figure 3
Figure 3
Finding the optimal alignment window size for ProQprof. Performance for neural nets trained using different sizes for the profile score window. “Profile only” refers to using only a window of profile similarity scores as input; “IC” and “gap” refer to the two last columns in the PSI-BLAST profile, respectively.
Figure 4
Figure 4
Schematic overview of the ProQprof and ProQres prediction schemes. ProQprof uses the target-template alignment for its prediction. Profiles are constructed for the target and template sequence. Profile-profile scores are calculated for aligned positions in the target-template alignment. The final prediction is done for the central residue in a window of profile-profile scores. ProQres analyzes the structure built from the target-template alignment. The prediction is based on the local structural environment around each residue described by structural features such as atom-atom and residue-residue contacts and surface accessibility (see Materials and Methods for details). Finally ProQres and ProQprof predictions are combined in ProQlocal using a simple sum of the two scores.
Figure 5
Figure 5
Performance comparison using ROC plots for methods using alignment information, i.e., ProQprof (thick), Profile-Profile window (thin), Profile-Profile no window (thick dotted), and Sequence-Profile window (thin dotted). The ability to find incorrectly and correctly aligned regions in both the LB2 (A and B, respectively) and the hmtest (C and D, respectively) sets was assessed.
Figure 6
Figure 6
Performance comparison using ROC plots for methods using structural information, i.e., ProQres (thin), ProsaII (thick), Verify3D (thick dotted), and Errat (thin dotted). The ability to find incorrectly and correctly aligned regions in both the LB2 (A and B, respectively) and the hmtest (C and D, respectively) sets was assessed.
Figure 7
Figure 7
Performance comparison using ROC plots for the best methods using alignment and/or structural information, i.e., Pcons-local (thick dotted), ProQlocal (thin dotted), ProQprof (thick), and ProQres (thin). The ability to find incorrectly and correctly aligned regions in both the LB2 (A and B, respectively) and the hmtest (C and D, respectively) sets was assessed.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Altschul S.F., Madden T.L., Schaffer A.A., Zhang J., Zhang Z., Miller W., Lipman D.J. 1997. Gapped BLAST and PSI-BLAST: A new generation of protein database search programs Nucleic Acids Res. 25: 3389–3402. - PMC - PubMed
    1. Bishop C.M. In Neural networks for pattern recognition. . 1995. Oxford University Press, Oxford,UK.
    1. Brunak S., Engelbrecht J., Knudsen S. 1991. Prediction of human mRNA donor and acceptor sites from the DNA sequence J. Mol. Biol. 220: 49–65. - PubMed
    1. Bujnicki J.M., Elofsson A., Fischer D., Rychlewski L. 2001. Live-bench-2: Large-scale automated evaluation of protein structure prediction servers Proteins 45: 184–191. - PubMed
    1. Colovos C. and Yeates T.O. 1993. Verification of protein structures: Patterns of nonbonded atomic interactions Protein Sci. 2: 1511–1519. - PMC - PubMed

Publication types

LinkOut - more resources

-