Local Geometry and Evolutionary Conservation of Protein Surfaces Reveal the Multiple Recognition Patches in Protein-Protein Interactions

Abstract

Protein-protein interactions (PPIs) are essential to all biological processes and they represent increasingly important therapeutic targets. Here, we present a new method for accurately predicting protein-protein interfaces, understanding their properties, origins and binding to multiple partners. Contrary to machine learning approaches, our method combines in a rational and very straightforward way three sequence- and structure-based descriptors of protein residues: evolutionary conservation, physico-chemical properties and local geometry. The implemented strategy yields very precise predictions for a wide range of protein-protein interfaces and discriminates them from small-molecule binding sites. Beyond its predictive power, the approach permits to dissect interaction surfaces and unravel their complexity. We show how the analysis of the predicted patches can foster new strategies for PPIs modulation and interaction surface redesign. The approach is implemented in JET2, an automated tool based on the Joint Evolutionary Trees (JET) method for sequence-based protein interface prediction. JET2 is freely available at www.lcqb.upmc.fr/JET2.

JET2 Viewer: a database of predicted multiple, possibly overlapping, protein–protein interaction sites for PDB structures

“…The data were generated by JET 2 , a computational method for the prediction of protein–protein binding sites that takes as input the 3D structure of a single protein and exploits both sequence and structural information. The combination of the geometric descriptor (circular variance) with sequence descriptors (conservation and physico–chemical properties of the residues) enables to capture with remarkable precision intrinsic features of protein binding patches (16). …”

“…Precise definitions of T JET , PC and CV are given in (16). The method to compute the evolutionary trace T JET and the propensity PC was first described in (8).…”

“…JET2 Viewer reports the three types of predictions (Figure 2), along with the results of JET 2 fully automated clustering algorithm. This algorithm detects interacting residues with the SC supporting the strongest interaction signal (depending on the system studied) and enriches the predictions with extra signal of interaction by using a complementary SC (16). …”

“…The predictions were produced by JET 2 , a new tool (16) that addresses the problem of identifying multiple interaction sites. The predictive model for protein interfaces implemented in JET 2 is inspired by a thorough analysis of known protein complexes which revealed a geometrical pattern observed for many binding sites (17).…”

“…A rational combination of these descriptors yields very precise predictions for a wide range of protein–protein interfaces and discriminates them from small-molecule binding sites. The method led us to go beyond the three-layer description and to highlight that interaction sites previously difficult to detect (8) are actually formed by either one or two layers (16). We could also identify interfaces shared by several partners, decrypt surfaces with several binding sites and decipher the evolutionary constraints that apply to different types of recognition patches.…”