Efficient Local Protein Structure Prediction

  • Szymon Nowakowski
  • Michał Drabikowski
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4481)


The methodology which was previously used with success in genomic sequences to predict new binding sites of transcription factors is applied in this paper for protein structure prediction. We predict local structure of proteins based on alignments of sequences of structurally similar local protein neighborhoods. We use Secondary Verification Assessment (SVA) method to select alignments with most reliable models. We show that using Secondary Verification (SV) method to assess the statistical significance of predictions we can reliably predict local protein structure, better than with the use of other methods (log-odds or p-value). The tests are conducted with the use of the test set consisting of the CASP 7 targets.


statistical significance SV method SVA method assessing predictions model assessment protein structure prediction CASP 7 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Nowakowski, S., Tiuryn, J.: A new approach to the assessment of the quality of predictions of transcription factor binding sites. Journal of Biomedical Informatics, in press (2006), doi:10.1016/j.jbi.2006.07.001Google Scholar
  2. 2.
    Hvidsten, T.R., et al.: A novel approach to fold recognition using sequence-derived properties from sets of structurally similar local fragments of proteins. Bioinformatics 19(Suppl. 2), II81–II91 (2003)Google Scholar
  3. 3.
    Drabikowski, M., Nowakowski, S., Tiuryn, J.: Library of local descriptors models the core of proteins accurately. Proteins: Structure, Function, and Bioinformatics, in press (2007)Google Scholar
  4. 4.
    Unger, R., et al.: A 3D building blocks approach to analyzing and predicting structure of proteins. Proteins: Struct. Funct. Genet. 5, 355–373 (1989)CrossRefGoogle Scholar
  5. 5.
    Yang, A., Wang, L.: Local structure prediction with local structure-based sequence profiles. Bioinformatics 19, 1267–1274 (2003)CrossRefGoogle Scholar
  6. 6.
    Sjölander, K., et al.: Dirichlet mixtures: a method for improved detection of weak but significant protein sequence homology. CABIOS 12, 327–345 (1996)Google Scholar
  7. 7.
    Brenner, S.E., Koehl, P., Levitt, M.: The ASTRAL compendium for protein structure and sequence analysis. Nucl. Acids Res. 28, 254–256 (2000)CrossRefGoogle Scholar
  8. 8.
    Drabikowski, M.: Analysis of groups and signals used to build protein structures from local descriptors (written in Polish). PhD thesis, Institute of Informatics, Warsaw University, Poland (2006)Google Scholar
  9. 9.
    Rahmann, S., Müller, T., Vingron, M.: On the power of profiles for transcription factor binding sites detection. Statistical Applications in Genetics and Molecular Biology 2(1) (2003)Google Scholar

Copyright information

© Springer Berlin Heidelberg 2007

Authors and Affiliations

  • Szymon Nowakowski
    • 1
  • Michał Drabikowski
    • 2
  1. 1.Infobright Inc., ul. Krzywickiego 34 pok. 219, 02-078 WarszawaPoland
  2. 2.Institute of Informatics, Warsaw UniversityPoland

Personalised recommendations