Real-SPINE: an integrated system of neural networks for real-value prediction of protein structural properties - PubMed

Affiliations

• PMID: 17397056
• DOI: 10.1002/prot.21408

Real-SPINE: an integrated system of neural networks for real-value prediction of protein structural properties

Ofer Dor et al. Proteins. 2007.

Abstract

Proteins can move freely in three-dimensional space. As a result, their structural properties, such as solvent accessible surface area, backbone dihedral angles, and atomic distances, are continuous variables. However, these properties are often arbitrarily divided into a few classes to facilitate prediction by statistical learning techniques. In this work, we establish an integrated system of neural networks (called Real-SPINE) for real-value prediction and apply the method to predict residue-solvent accessibility and backbone psi dihedral angles of proteins based on information derived from sequences only. Real-SPINE is trained with a large data set of 2640 protein chains, sequence profiles generated from multiple sequence alignment, representative amino-acid properties, a slow learning rate, overfitting protection, and predicted secondary structures. The method optimizes more than 200,000 weights and yields a 10-fold cross-validated Pearson's correlation coefficient (PCC) of 0.74 between predicted and actual solvent accessible surface areas and 0.62 between predicted and actual psi angles. In particular, 90% of 2640 proteins have a PCC value greater than 0.6 between predicted and actual solvent-accessible surface areas. The results of Real-SPINE can be compared with the best reported correlation coefficients of 0.64-0.67 for solvent-accessible surface areas and 0.47 for psi angles. The real-SPINE server, executable programs, and datasets are freely available on http://sparks.informatics.iupui.edu.

Similar articles

• Achieving 80% ten-fold cross-validated accuracy for secondary structure prediction by large-scale training.

  Dor O, Zhou Y. Dor O, et al. Proteins. 2007 Mar 1;66(4):838-45. doi: 10.1002/prot.21298. Proteins. 2007. PMID: 17177203

• Real value prediction of solvent accessibility in proteins using multiple sequence alignment and secondary structure.

  Garg A, Kaur H, Raghava GP. Garg A, et al. Proteins. 2005 Nov 1;61(2):318-24. doi: 10.1002/prot.20630. Proteins. 2005. PMID: 16106377

• Real-value prediction of backbone torsion angles.

  Xue B, Dor O, Faraggi E, Zhou Y. Xue B, et al. Proteins. 2008 Jul;72(1):427-33. doi: 10.1002/prot.21940. Proteins. 2008. PMID: 18214956

• Accurate prediction of solvent accessibility using neural networks-based regression.

  Adamczak R, Porollo A, Meller J. Adamczak R, et al. Proteins. 2004 Sep 1;56(4):753-67. doi: 10.1002/prot.20176. Proteins. 2004. PMID: 15281128

• SVM-Cabins: prediction of solvent accessibility using accumulation cutoff set and support vector machine.

  Wang JY, Lee HM, Ahmad S. Wang JY, et al. Proteins. 2007 Jul 1;68(1):82-91. doi: 10.1002/prot.21422. Proteins. 2007. PMID: 17436325

Cited by

• Protein Solvent-Accessibility Prediction by a Stacked Deep Bidirectional Recurrent Neural Network.

  Zhang B, Li L, Lü Q. Zhang B, et al. Biomolecules. 2018 May 25;8(2):33. doi: 10.3390/biom8020033. Biomolecules. 2018. PMID: 29799510 Free PMC article.

• CPHmodels-3.0--remote homology modeling using structure-guided sequence profiles.

  Nielsen M, Lundegaard C, Lund O, Petersen TN. Nielsen M, et al. Nucleic Acids Res. 2010 Jul;38(Web Server issue):W576-81. doi: 10.1093/nar/gkq535. Epub 2010 Jun 11. Nucleic Acids Res. 2010. PMID: 20542909 Free PMC article.

• Molecular characterization of the apoptosis-related SH3RF1 and SH3RF2 genes and their association with exercise performance in Arabian horses.

  Ropka-Molik K, Stefaniuk-Szmukier M, Piórkowska K, Szmatoła T, Bugno-Poniewierska M. Ropka-Molik K, et al. BMC Vet Res. 2018 Aug 14;14(1):237. doi: 10.1186/s12917-018-1567-0. BMC Vet Res. 2018. PMID: 30107803 Free PMC article.

• SP5: improving protein fold recognition by using torsion angle profiles and profile-based gap penalty model.

  Zhang W, Liu S, Zhou Y. Zhang W, et al. PLoS One. 2008 Jun 4;3(6):e2325. doi: 10.1371/journal.pone.0002325. PLoS One. 2008. PMID: 18523556 Free PMC article.

• GENN: a GEneral Neural Network for learning tabulated data with examples from protein structure prediction.

  Faraggi E, Kloczkowski A. Faraggi E, et al. Methods Mol Biol. 2015;1260:165-78. doi: 10.1007/978-1-4939-2239-0_10. Methods Mol Biol. 2015. PMID: 25502381 Free PMC article.

Publication types

MeSH terms

Substances

LinkOut - more resources

• Full Text Sources

  • Wiley

• Other Literature Sources

  • The Lens - Patent Citations Database

• Research Materials

  • NCI CPTC Antibody Characterization Program