A DNN Biophysics Model with Topological and Electrostatic Features

2024-09-05

arXiv.org

AI Summary

This study addresses the challenge of improving prediction accuracy for protein biophysical properties—specifically, electrostatic solvation energy. We propose a deep neural network (DNN) framework that jointly encodes multiscale topological and electrostatic features: for the first time, element-specific persistent homology (ESPH) captures structural topology, while Cartesian tree-code–accelerated electrostatic potential features encode force-field information; both representations are scale-invariant and scalable, enabling unified characterization of protein structure and physical interactions. Evaluated on over 4,000 protein structures, the dual-feature integration significantly outperforms single-feature baselines, yielding substantial gains in prediction accuracy. The framework supports cross-scale modeling and efficient training on large-scale structural databases. Moreover, its design principles are generalizable to other biomacromolecules—including nucleic acids and glycans—offering broad applicability in computational biophysics.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Evolutionary Computation
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Sustainability of Web economics
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

In this project, we provide a deep-learning neural network (DNN) based biophysics model to predict protein properties. The model uses multi-scale and uniform topological and electrostatic features generated with protein structural information and force field, which governs the molecular mechanics. The topological features are generated using the element specified persistent homology (ESPH) while the electrostatic features are fast computed using a Cartesian treecode. These features are uniform in number for proteins with various sizes thus the broadly available protein structure database can be used in training the network. These features are also multi-scale thus the resolution and computational cost can be balanced by the users. The machine learning simulation on over 4000 protein structures shows the efficiency and fidelity of these features in representing the protein structure and force field for the predication of their biophysical properties such as electrostatic solvation energy. Tests on topological or electrostatic features alone and the combination of both showed the optimal performance when both features are used. This model shows its potential as a general tool in assisting biophysical properties and function prediction for the broad biomolecules using data from both theoretical computing and experiments.

Problem

• Predict protein properties using DNN with topological and electrostatic features
• Balance resolution and computational cost with multi-scale features
• Assist biophysical property prediction for broad biomolecules

Innovation

• DNN model with topological and electrostatic features
• Multi-scale uniform features for various protein sizes
• Combined features enhance biophysical property prediction