Protein-protein interface hot spots prediction based on a hybrid feature selection strategy

Abstract

BackgroundHot spots are interface residues that contribute most binding affinity to protein-protein interaction. A compact and relevant feature subset is important for building machine learning methods to predict hot spots on protein-protein interfaces. Although different methods have been used to detect the relevant feature subset from a variety of features related to interface residues, it is still a challenge to detect the optimal feature subset for building the final model.ResultsIn this study, three different feature selection methods were compared to propose a new hybrid feature selection strategy. This new strategy was proved to effectively reduce the feature space when we were building the prediction models for identifying hotspot residues. It was tested on eighty-two features, both conventional and newly proposed. According to the strategy, combining the feature subsets selected by decision tree and mRMR (maximum Relevance Minimum Redundancy) individually, we were able to build a model with 6 features by using a PSFS (Pseudo Sequential Forward Selection) process. Compared with other state-of-art methods for the independent test set, our model had shown better or comparable predictive performances (with F-measure 0.622 and recall 0.821). Analysis of the 6 features confirmed that our newly proposed feature CNSV_REL1 was important for our model. The analysis also showed that the complementarity between features should be considered as an important aspect when conducting the feature selection.ConclusionIn this study, most important of all, a new strategy for feature selection was proposed and proved to be effective in selecting the optimal feature subset for building prediction models, which can be used to predict hot spot residues on protein-protein interfaces. Moreover, two aspects, the generalization of the single feature and the complementarity between features, were proved to be of great importance and should be considered in feature selection methods. Finally, our newly proposed feature CNSV_REL1 had been proved an alternative and effective feature in predicting hot spots by our study. Our model is available for users through a webserver: http://zhulab.ahu.edu.cn/iPPHOT/.Electronic supplementary materialThe online version of this article (10.1186/s12859-018-2009-5) contains supplementary material, which is available to authorized users.

Hot-spots and their contribution to the self-assembly of the viral capsid: in-silico prediction and analysis

“…Comparison with other hot-spot prediction tools was made. We used the energy-based alanine scanning mutagenesis computational methods implemented in the ROBBETA [1], FoldX [2], SpotOn [3], and iPPHOT [4] (alignment created by ConSurf [25] using UNIREF90 and MAFFT) online tools. These are fast but coarse approaches for the prediction of energetically relevant amino acid residues in protein-protein interfaces.…”

“…Figure A2. iPPHOT [4] hot-spot predictions for the structure of a representative 2-fold related dimer. All interface residues were predicted to be null-spots (NS).…”

“…Usually, an energy-based alanine scanning mutagenesis method is implemented for this purpose. Examples of this type of predictors are ROBBETA [1], FoldX [2], SpotOn [3], and iPPHOT [4]. Other approaches have opted for machine-learning-based methods [5][6][7][8], molecular-dynamics-based methods [9,10], or combining solvent accessibility and inter-residue potentials [11].…”

Prediction of quaternary structure by analysis of hot spot residues in protein‐protein interfaces: the case of anthranilate phosphoribosyltransferases

“…By definition, a residue in a protein‐protein interface is termed hot spot if its mutation to alanine changes the free binding energy of the interaction substantially (ΔΔ G binding ≥ 2.0 kcal/mol). Since experimental alanine scanning is an expensive and time‐consuming procedure that is not applicable on a large scale, highly efficient in silico methods, often based on machine learning approaches, have been developed in order to predict hot spots from the native complex structure or even from the sequence of one of the binding partners . Feature‐based prediction approaches use a variety of different chemical and physical characteristics of interface residues, such as solvent accessible surface area, protrusion index, residue conservation, or B factors.…”

“…Since experimental alanine scanning is an expensive and time-consuming procedure that is not applicable on a large scale, highly efficient in silico methods, often based on machine learning approaches, have been developed in order to predict hot spots from the native complex structure or even from the sequence of one of the binding partners. [23][24][25][26] Featurebased prediction approaches use a variety of different chemical and physical characteristics of interface residues, such as solvent accessible surface area, protrusion index, residue conservation, or B factors. However, predictive performances of different methods vary with data sets, and experimental validations of those computational predictions are scarce.…”