Classification comparison of prediction of solvent accessibility from protein sequences

ABSTRACT

The prediction of residue solvent accessibility from protein sequences has been studied by various methods. The direct comparison of these methods is impossible due to the variety of datasets used and the difference in structure definition. In this paper we choose 5 classification approaches (decision tree (DT), Support Vector Machine (SVM), Bayesian Statistics (BS), Neural Network (NN) and Multiple Linear Regression (MLR)) for predicting solvent accessibility based on the same dataset and using the same structure definition so that we can directly compare different methods. We evaluate these methods in a cross-validation test on 2148 unique proteins using single sequences and multiple sequences approaches with a cutoff of 20% for two-state definition of solvent accessibility. According to the experiment results, SVM and NN are both the best predictors with accuracy 79%, correlation coefficient 0.59, 2~4% superior to other three methods on multiple sequences prediction. A further test result on a blind test set from Critical Assessment of Techniques for Protein Structure Prediction experiment (CASP5) is consistent with this result. On single sequence prediction, DT, BS and MLR perform about the same at 71~72% with correlation coefficient 0.43. The improvement over the baseline model that use only the identity of target residue is small. Local sequence seems embed very little information on accessibility. Separate training according to protein size improves the prediction when there are sufficiently large dataset available. The consensus prediction combining the 5 approaches is not significantly better than the best single method.

References

1. Altschul, Stephen F., Thomas L. Madden, Alejandro A. Schaffer, Jinghui Zhang, Zheng Zhang, Webb Miller, and David J. Lipman (1997), Nucleic Acids Res. 25:3389--3402.
2. Andrade M. A., O'Donoghue S. I., Rost B. Adaptation of protein surfaces to subcellular location. J. Mol. Biol. 1998; 276:517--525.
3. Anfinsen C. B. Principles that govern the folding of protein chains. Science 181:223--230, 1973.
4. Chan, H. S. and Dill, K. A. (1990) Proc. Natl. Acad. Sci. USA, 87, 6388--6392.
5. Delamarche C., Guerdoux-Jamet P., Gras R., Nicolas J. Biochimie 81: 1065--1072 (1999).
6. Ding CHQ, Dubchak I. Bioinformatics 2001; 17:349--358.
7. Ehrlich, L., Reczko, M., Bohr, H. and Wade, R. C. (1998) Protein Engng, 11, 11--19.
8. Furey T. S., Cristianini N., Duffy N., Bednarski D. W., Schummer M., Haussler D. Bioinformatics 2000; 16:906--914.
9. Holm L. and Sander C. (1996) Mapping the protein universe. Science 273:595--602.
10. Horton P., Nakai K. Intelligent Systems in Molecular Biology 5:147--152(1997).
11. Hua S., Sun Z. J Mol Biol 2001;308:397--407.
12. Joachims T., Making large-Scale SVM Learning Practical. Advances in Kernel Methods - Support Vector Learning, B. Schölkopf and C. Burges and A. Smola (ed.), MIT-Press, 1999.
13. Jones D. Protein secondary structure prediction based on position-specific scoring matrices, J. Mol. Biol (1999); 292:195--202.
14. Kabsch W., Sander C. Dictionary of protein secondary structures: pattern recognition of hydrogen bonded and geometrical features. Biopolymers 22: 2577--2637 (1983).
15. Kyte J. & Doolittle, R. F. (1982) J. Mol. Biol. 157, 105--132.
16. Li X., Pan X-M. New methods for accurate prediction of solvent accessibility from protein sequence. Proteins 42:1--5 (2001)
17. Naderi-Manesh H., Sadeghi M., Arab S., Movahedi AAM. Prediction of protein surface accessibility with information theory. Proteins 42: 452--459 (2001).
18. Nelson D., Cox M. Lehninger Principles of Biochemistry (3rd ed.) Page 118.
19. Ooi T., Oobatake M., Nemethy G., Scheraga H. A., (1987) Accessible suface areas as a measure of the thermodynamics parameters of hydration of peptides. Proc. Natl. Acad. Sci USA 84:3086--3090.
20. Richardson C. J., Barlow D. J. The bottom line for prediction of residue solvent accessibility. Protein Eng 12: 1051--1054 (1999).
21. Rost B., and Sander C: Conservation and prediction of solvent accessibility in protein families. Proteins, 20, 216--226, 1994.
22. Rost B., Sander C. Progress of 1D protein structure prediction at last. Proteins 23:295--300 (1995).
23. Salzberg S. L., Delcher A. L., Fasman K. H., Henderson J. A decision tree system for finding genes in DNA. J. Comput. Biol. 5: 667--680 (1998).
24. Selbig J., Mevissen T., Lengauer T. Decision tree-based formation of consensus protein secondary structure prediction. Bioinformatics 15: 1039--1046 (1999)
25. Shan Y., Wang G., Zhou H. Fold recognition and accurate query-template alignment by a combination of Psi-Blast and threading. Proteins 42: 23--37 (2001).
26. Shrake A., Rupley JA. Environment and exposure to solvent of protein atoms: lysozyme and insulin. J Mol Biol 79: 351--371 (1973).
27. Thompson M. J., Goldstein RA. Predicting solvent accessibility: higher accuracy using Bayesian statistics and optimized residue substitution classes. Proteins 25: 38--47 (1996).
28. Varadarajan R., Nagarajaram H. A. and Ramakrishnan C(1996) Proc. Natl. Acad. Sci. USA 93, 13908--13913.
29. Yuan Z., Burrage K., Mattick J. Prediction of protein solvent accessibility using support vector machines. Proteins 48:566--570 (2002).

Index Terms

• Applied computing

  • Life and medical sciences

    • Computational biology

    • Genetics

    • Systems biology

• Computer systems organization

  • Architectures

    • Other architectures

      • Neural networks

• Computing methodologies

  • Machine learning

    • Learning paradigms

      • Supervised learning

        • Supervised learning by classification

    • Machine learning approaches

      • Classification and regression trees

      • Neural networks