Complementing sequence-derived features with structural information extracted from fragment libraries for protein structure prediction

Siyuan Liu,Qijiang Xu,Jian Yin,Tong Wang,Tie-Yan Liu,Bin Shao

doi:10.1186/s12859-021-04258-6

Abstract

BackgroundFragment libraries play a key role in fragment-assembly based protein structure prediction, where protein fragments are assembled to form a complete three-dimensional structure. Rich and accurate structural information embedded in fragment libraries has not been systematically extracted and used beyond fragment assembly.MethodsTo better leverage the valuable structural information for protein structure prediction, we extracted seven types of structural information from fragment libraries. We broadened the usage of such structural information by transforming fragment libraries into protein-specific potentials for gradient-descent based protein folding and encoding fragment libraries as structural features for protein property prediction.ResultsFragment libraires improved the accuracy of protein folding and outperformed state-of-the-art algorithms with respect to predicted properties, such as torsion angles and inter-residue distances.ConclusionOur work implies that the rich structural information extracted from fragment libraries can complement sequence-derived features to help protein structure prediction.

Highlights

Fragment libraries play a key role in fragment-assembly based protein structure prediction, where protein fragments are assembled to form a complete three-dimensional structure
Rich and accurate structural information in fragment libraries Initially designed to serve in fragment assembly-based simulations, fragment libraries are lists of short template structures that are considered resemble to continuous regions of the structure of a target protein
A fragment library consists of thousands of fragments and contains rich structural information about the target protein, such information is lack of exploration and seldom used beyond fragment assembly

Summary

Introduction

Fragment libraries play a key role in fragment-assembly based protein structure prediction, where protein fragments are assembled to form a complete three-dimensional structure. Rich and accurate structural information embedded in fragment libraries has not been systematically extracted and used beyond fragment assembly. Protein structure prediction is one of the most challenging and active research fields in computational biology [1,2,3]. The high-quality fragment libraries are one of the determining factors to the success of fragment assembly [8, 9]. Fragment libraries contain rich structural information, including 1D structural properties such as secondary structure and torsion angles, and 2D structural properties such as distances and orientations between pairs of heavy atoms. Fragment libraries are extensively utilized in fragment assembly, the rich structural information has not yet been systematically analyzed and leveraged by other protein structure prediction approaches

Methods

Results

Conclusion