Abstract:The prediction of 3D structures of proteins from amino acid sequences is one of the most challenging problems in molecular biology. An essential task for solving this problem with coarse-grained models is to deduce effective interaction potentials. The development and evaluation of new energy functions is critical to accurately modeling the properties of biological macromolecules. Knowledge-based mean force potentials are derived from statistical analysis of proteins of known structures. Current knowledge-based potentials are almost in the form of weighted linear sum of interaction pairs. In this study, a class of novel nonlinear knowledge-based mean force potentials is presented. The potential parameters are obtained by nonlinear classifiers, instead of relative frequencies of interaction pairs against a reference state or linear classifiers. The support vector machine is used to derive the potential parameters on data sets that contain both native structures and decoy structures. Five knowledge-based mean force Boltzmann-based or linear potentials are introduced and their corresponding nonlinear potentials are implemented. They are the DIH potential (single-body residue-level Boltzmann-based potential), the DFIRE-SCM potential (two-body residue-level Boltzmann-based potential), the FS potential (two-body atom-level Boltzmann-based potential), the HR potential (two-body residue-level linear potential), and the T32S3 potential (two-body atom-level linear potential). Experiments are performed on well-established decoy sets, including the LKF data set, the CASP7 data set, and the Decoys "R" Us data set. The evaluation metrics include the energy Z score and the ability of each potential to discriminate native structures from a set of decoy structures. Experimental results show that all nonlinear potentials significantly outperform the corresponding Boltzmann-based or linear potentials, and the proposed discriminative framework is effective in developing knowledge-based mean force potentials. The nonlinear potentials can be widely used for ab initio protein structure prediction, model quality assessment, protein docking, and other challenging problems in computational biology.

Knowledge-based energy functions for computational studies of proteins

Novel Nonlinear Knowledge-Based Mean Force Potentials Based On Machine Learning

Empirical Probability Functions Derived from Dihedral Angles for Protein Structure Prediction

An All‐atom Knowledge‐based Energy Function for Protein‐dna Threading, Docking Decoy Discrimination, and Prediction of Transcription‐factor Binding Profiles

A Discontinuous Potential Model for Protein-Protein Interactions.

Designing of knowledge-based potentials via B-spline basis functions for native proteins detection

ADVANCES IN RESEARCH ON RESIDUE INTERACTION POTENTIAL ENERGY FUNCTION BASED ON PROTEIN CRYSTAL STRUCTURES

Potentials of Mean Force for Protein Structure Prediction Vindicated, Formalized and Generalized

Potential of mean force for protein-protein interaction studies.

A Statistical Analysis of Protein-Protein Interaction with Knowledge-Based Potential at Residue Level

Pair potentials for protein folding: Choice of reference states and sensitivity of predicted native states to variations in the interaction schemes

Computational prediction of protein–protein binding affinities

A Novel Score for Highly Accurate and Efficient Prediction of Native Protein Structures

A New-Protein Folding Recognition Potential Function

Learning effective amino acid interactions through iterative stochastic techniques

Quantitative Evaluation of Protein-Dna Interactions Using an Optimized Knowledge-Based Potential.

Chapter 4. Predicting and characterizing protein functions through matching geometric and evolutionary patterns of binding surfaces.

Statistical potential for assessment and prediction of protein structures

Protein-Protein Interactions: Interface Analysis, Binding Free Energy Calculation and Interaction Design

An Improved Genetic Algorithm for Statistical Potential Function Design and Protein Structure Prediction

Incorporating a multiobjective knowledge-based energy function into differential evolution for protein structure prediction