Toward an accurate prediction of inter-residue distances in proteins using 2D recursive neural networks

Hdl Handle:
http://hdl.handle.net/10147/315472
Title:
Toward an accurate prediction of inter-residue distances in proteins using 2D recursive neural networks
Authors:
Kukic, Predrag; Mirabello, Claudio; Tradigo, Giuseppe; Walsh, Ian; Veltri, Pierangelo; Pollastri, Gianluca
Citation:
BMC bioinformatics. 2014 Jan 10;15(1):6
Journal:
BMC bioinformatics
Issue Date:
10-Jan-2014
URI:
http://dx.doi.org/10.1186/1471-2105-15-6; http://hdl.handle.net/10147/315472
Abstract:
Abstract Background Protein inter-residue contact maps provide a translation and rotation invariant topological representation of a protein. They can be used as an intermediary step in protein structure predictions. However, the prediction of contact maps represents an unbalanced problem as far fewer examples of contacts than non-contacts exist in a protein structure.In this study we explore the possibility of completely eliminating the unbalanced nature of the contact map prediction problem by predicting real-value distances between residues. Predicting full inter-residue distance maps and applying them in protein structure predictions has been relatively unexplored in the past. Results We initially demonstrate that the use of native-like distance maps is able to reproduce 3D structures almost identical to the targets, giving an average RMSD of 0.5Å. In addition, the corrupted physical maps with an introduced random error of ±6Å are able to reconstruct the targets within an average RMSD of 2Å.After demonstrating the reconstruction potential of distance maps, we develop two classes of predictors using two-dimensional recursive neural networks: an ab initio predictor that relies only on the protein sequence and evolutionary information, and a template-based predictor in which additional structural homology information is provided. We find that the ab initio predictor is able to reproduce distances with an RMSD of 6Å, regardless of the evolutionary content provided. Furthermore, we show that the template-based predictor exploits both sequence and structure information even in cases of dubious homology and outperforms the best template hit with a clear margin of up to 3.7Å.Lastly, we demonstrate the ability of the two predictors to reconstruct the CASP9 targets shorter than 200 residues producing the results similar to the state of the machine learning art approach implemented in the Distill server. Conclusions The methodology presented here, if complemented by more complex reconstruction protocols, can represent a possible path to improve machine learning algorithms for 3D protein structure prediction. Moreover, it can be used as an intermediary step in protein structure predictions either on its own or complemented by NMR restraints.
Item Type:
Article
Language:
en
Keywords:
GENETICS
Local subject classification:
NEUROLOGY

Full metadata record

DC FieldValue Language
dc.contributor.authorKukic, Predragen_GB
dc.contributor.authorMirabello, Claudioen_GB
dc.contributor.authorTradigo, Giuseppeen_GB
dc.contributor.authorWalsh, Ianen_GB
dc.contributor.authorVeltri, Pierangeloen_GB
dc.contributor.authorPollastri, Gianlucaen_GB
dc.date.accessioned2014-04-07T11:15:37Z-
dc.date.available2014-04-07T11:15:37Z-
dc.date.issued2014-01-10-
dc.identifier.citationBMC bioinformatics. 2014 Jan 10;15(1):6en_GB
dc.identifier.urihttp://dx.doi.org/10.1186/1471-2105-15-6-
dc.identifier.urihttp://hdl.handle.net/10147/315472-
dc.description.abstractAbstract Background Protein inter-residue contact maps provide a translation and rotation invariant topological representation of a protein. They can be used as an intermediary step in protein structure predictions. However, the prediction of contact maps represents an unbalanced problem as far fewer examples of contacts than non-contacts exist in a protein structure.In this study we explore the possibility of completely eliminating the unbalanced nature of the contact map prediction problem by predicting real-value distances between residues. Predicting full inter-residue distance maps and applying them in protein structure predictions has been relatively unexplored in the past. Results We initially demonstrate that the use of native-like distance maps is able to reproduce 3D structures almost identical to the targets, giving an average RMSD of 0.5Å. In addition, the corrupted physical maps with an introduced random error of ±6Å are able to reconstruct the targets within an average RMSD of 2Å.After demonstrating the reconstruction potential of distance maps, we develop two classes of predictors using two-dimensional recursive neural networks: an ab initio predictor that relies only on the protein sequence and evolutionary information, and a template-based predictor in which additional structural homology information is provided. We find that the ab initio predictor is able to reproduce distances with an RMSD of 6Å, regardless of the evolutionary content provided. Furthermore, we show that the template-based predictor exploits both sequence and structure information even in cases of dubious homology and outperforms the best template hit with a clear margin of up to 3.7Å.Lastly, we demonstrate the ability of the two predictors to reconstruct the CASP9 targets shorter than 200 residues producing the results similar to the state of the machine learning art approach implemented in the Distill server. Conclusions The methodology presented here, if complemented by more complex reconstruction protocols, can represent a possible path to improve machine learning algorithms for 3D protein structure prediction. Moreover, it can be used as an intermediary step in protein structure predictions either on its own or complemented by NMR restraints.-
dc.language.isoenen
dc.subjectGENETICSen_GB
dc.subject.otherNEUROLOGYen_GB
dc.titleToward an accurate prediction of inter-residue distances in proteins using 2D recursive neural networksen_GB
dc.typeArticleen
dc.identifier.journalBMC bioinformaticsen_GB
dc.language.rfc3066en-
dc.rights.holderPredrag Kukic et al.; licensee BioMed Central Ltd.-
dc.description.statusPeer Reviewed-
dc.date.updated2014-04-02T10:41:48Z-
All Items in Lenus, The Irish Health Repository are protected by copyright, with all rights reserved, unless otherwise indicated.