Improving Predictions of Protein-Protein Interfaces by Combining Amino Acid-Specific Classifiers Based on Structural and Physicochemical Descriptors with Their Weighted Neighbor Averages

Gajendra P S Raghava,Fábio R De Moraes,José G Jardine,José A Salim,Goran Neshich,Ivan Mazoni,Izabella A P Neshich,Inácio H Yano,José G C Pereira

doi:10.1371/journal.pone.0087107

Gajendra P S Raghava, Fábio R De Moraes + Show 7 more

Open Access

https://doi.org/10.1371/journal.pone.0087107

Copy DOI

Abstract

Protein-protein interactions are involved in nearly all regulatory processes in the cell and are considered one of the most important issues in molecular biology and pharmaceutical sciences but are still not fully understood. Structural and computational biology contributed greatly to the elucidation of the mechanism of protein interactions. In this paper, we present a collection of the physicochemical and structural characteristics that distinguish interface-forming residues (IFR) from free surface residues (FSR). We formulated a linear discriminative analysis (LDA) classifier to assess whether chosen descriptors from the BlueStar STING database (http://www.cbi.cnptia.embrapa.br/SMS/) are suitable for such a task. Receiver operating characteristic (ROC) analysis indicates that the particular physicochemical and structural descriptors used for building the linear classifier perform much better than a random classifier and in fact, successfully outperform some of the previously published procedures, whose performance indicators were recently compared by other research groups. The results presented here show that the selected set of descriptors can be utilized to predict IFRs, even when homologue proteins are missing (particularly important for orphan proteins where no homologue is available for comparative analysis/indication) or, when certain conformational changes accompany interface formation. The development of amino acid type specific classifiers is shown to increase IFR classification performance. Also, we found that the addition of an amino acid conservation attribute did not improve the classification prediction. This result indicates that the increase in predictive power associated with amino acid conservation is exhausted by adequate use of an extensive list of independent physicochemical and structural parameters that, by themselves, fully describe the nano-environment at protein-protein interfaces. The IFR classifier developed in this study is now integrated into the BlueStar STING suite of programs. Consequently, the prediction of protein-protein interfaces for all proteins available in the PDB is possible through STING_interfaces module, accessible at the following website: (http://www.cbi.cnptia.embrapa.br/SMS/predictions/index.html).

Highlights

Protein-protein interactions are very specific in the sense that they control almost all processes within cells, such as signal transduction, metabolic and gene regulation, and immunologic responses. [1]
We can observe that the attributes Contact Energy Density (CED) at the Ca (6A), CED at Last Heavy Atom (LHA) (5A), CED at LHA (6A), density at Ca (6A), density at LHA (6A), sponge at Ca (6A), sponge at LHA (6A), electrostatic potential-average and Cross Link Order at Cb are linearly correlated with another variation of the same descriptor in most amino acid types
Our results indicate that all the necessary information for distinguishing interface-forming residues (IFR) from free surface residues (FSR) is present in the original descriptor set, if a sufficiently extensive list is used

Summary

Introduction

Protein-protein interactions are very specific in the sense that they control almost all processes within cells, such as signal transduction, metabolic and gene regulation, and immunologic responses. [1]. Similar to some experimental techniques used to gather information about protein interfaces (for example, obtaining 3D protein structures through x-ray diffraction or nuclear magnetic resonance techniques), computational methods do face some challenges. These difficulties include predicting quaternary structure via template-based docking algorithms, which can only yield atomic details of protein-protein interactions if the sequence identity to another known protein complex structure is higher than approximately 60%. An advanced knowledge of this particular location may result in much better structure predictions of the entire complex This improvement is mainly because all protein-protein interactions occur only at a portion of the protein surface: the interface between the molecules. It has been argued that monomeric subunits have all the necessary features for establishing proteinprotein interactions [1]

Methods

Results

Conclusion