A Scalable and Quantum-Accurate Foundation Model for Biomolecular Force Field via Linearly Tensorized Quadrangle Attention

2025-07-01

AI Summary

Atomic-scale biomolecular simulation is hindered by the dual bottlenecks of insufficient accuracy in classical force fields and prohibitive computational cost of quantum mechanical methods. Existing AI-based force fields (AIFFs) struggle to simultaneously achieve many-body expressivity, broad transferability, and computational efficiency. To address this, we propose LiTEN—a rotationally equivariant neural network force field—featuring a novel tensorized quadrilateral attention (TQA) mechanism with linear complexity that bypasses spherical harmonic computations while efficiently reconstructing high-order tensor features. LiTEN further incorporates vectorized reparameterization and a two-stage training strategy: nablaDFT pretraining followed by SPICE fine-tuning. On benchmarks including rMD17, MD22, and Chignolin, LiTEN achieves state-of-the-art accuracy. For biomolecules with ~1,000 atoms, it delivers 10× faster inference than MACE-OFF, enabling scalable conformational sampling, geometry optimization, and free energy surface construction.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Sampling/Simulation-based Search
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Accurate atomistic biomolecular simulations are vital for disease mechanism understanding, drug discovery, and biomaterial design, but existing simulation methods exhibit significant limitations. Classical force fields are efficient but lack accuracy for transition states and fine conformational details critical in many chemical and biological processes. Quantum Mechanics (QM) methods are highly accurate but computationally infeasible for large-scale or long-time simulations. AI-based force fields (AIFFs) aim to achieve QM-level accuracy with efficiency but struggle to balance many-body modeling complexity, accuracy, and speed, often constrained by limited training data and insufficient validation for generalizability. To overcome these challenges, we introduce LiTEN, a novel equivariant neural network with Tensorized Quadrangle Attention (TQA). TQA efficiently models three- and four-body interactions with linear complexity by reparameterizing high-order tensor features via vector operations, avoiding costly spherical harmonics. Building on LiTEN, LiTEN-FF is a robust AIFF foundation model, pre-trained on the extensive nablaDFT dataset for broad chemical generalization and fine-tuned on SPICE for accurate solvated system simulations. LiTEN achieves state-of-the-art (SOTA) performance across most evaluation subsets of rMD17, MD22, and Chignolin, outperforming leading models such as MACE, NequIP, and EquiFormer. LiTEN-FF enables the most comprehensive suite of downstream biomolecular modeling tasks to date, including QM-level conformer searches, geometry optimization, and free energy surface construction, while offering 10x faster inference than MACE-OFF for large biomolecules (~1000 atoms). In summary, we present a physically grounded, highly efficient framework that advances complex biomolecular modeling, providing a versatile foundation for drug discovery and related applications.

Problem

• Classical force fields lack accuracy for critical biomolecular states.
• QM methods are too slow for large-scale biomolecular simulations.
• AI force fields struggle with complexity, accuracy, and speed balance.

Innovation

• Equivariant neural network with Tensorized Quadrangle Attention
• Linear complexity for three- and four-body interactions
• Pre-trained on nablaDFT and fine-tuned on SPICE