A Novel Geometry-Based Approach to Infer Protein Interface Similarity

Inbal Budowski-Tal,Rachel Kolodny,Yael Mandel-Gutfreund

doi:10.1038/s41598-018-26497-z

Inbal Budowski-Tal, Rachel Kolodny + Show 1 more

Open Access

https://doi.org/10.1038/s41598-018-26497-z

Copy DOI

Abstract

The protein interface is key to understand protein function, providing a vital insight on how proteins interact with each other and with other molecules. Over the years, many computational methods to compare protein structures were developed, yet evaluating interface similarity remains a very difficult task. Here, we present PatchBag – a geometry based method for efficient comparison of protein surfaces and interfaces. PatchBag is a Bag-Of-Words approach, which represents complex objects as vectors, enabling to search interface similarity in a highly efficient manner. Using a novel framework for evaluating interface similarity, we show that PatchBag performance is comparable to state-of-the-art alignment-based structural comparison methods. The great advantage of PatchBag is that it does not rely on sequence or fold information, thus enabling to detect similarities between interfaces in unrelated proteins. We propose that PatchBag can contribute to reveal novel evolutionary and functional relationships between protein interfaces.

Highlights

The protein interface is key to understand protein function, providing a vital insight on how proteins interact with each other and with other molecules
We propose that PatchBag can help to identify similar interfaces regardless of their fold, providing a novel approach for protein function prediction based on their functional interfaces
The protein surface is represented by all protein residues that are exposed to the solvent, while interfaces are defined as the subset of residue that interact with a specific partner (See Methods)

Summary

Introduction

The protein interface is key to understand protein function, providing a vital insight on how proteins interact with each other and with other molecules. Using a novel framework for evaluating interface similarity, we show that PatchBag performance is comparable to state-ofthe-art alignment-based structural comparison methods. Several algorithms were developed to identify surface similarities, independent of the overall protein folds These algorithms represent the surface shapes in various ways, such as Alpha Shapes and Delaunay Triangulations[15,16,17], Three-Dimensional Zernike Descriptors (3DZD)[18,19,20,21], or an unordered collection of the three-dimensional (3D) coordinates of the surface atoms[22,23]. Other methods based on local structural surface comparison have been employed for pocket comparison[42] and interface prediction, for instance, predicting protein-protein interacting residues[43] or finding ligand binding sites[21,44].

Methods

Results

Conclusion