Mamba-driven multi-perspective structural understanding for molecular ground-state conformation prediction

TL;DR

This paper introduces MPSU-Mamba, a novel framework leveraging the Mamba model for multi-perspective understanding of molecular structures to accurately predict ground-state conformations, especially effective with limited training data.

Contribution

The paper proposes a generic Mamba-driven framework with multi-perspective strategies and a bright-channel mechanism for improved molecular conformation prediction.

Findings

Outperforms existing methods on QM9 and Molecule3D datasets.

Effective even with few training samples.

Provides comprehensive structural understanding of molecules.

Abstract

A comprehensive understanding of molecular structures is important for the prediction of molecular ground-state conformation involving property information. Meanwhile, state space model (e.g., Mamba) has recently emerged as a promising mechanism for long sequence modeling and has achieved remarkable results in various language and vision tasks. However, towards molecular ground-state conformation prediction, exploiting Mamba to understand molecular structure is underexplored. To this end, we strive to design a generic and efficient framework with Mamba to capture critical components. In general, molecular structure could be considered to consist of three elements, i.e., atom types, atom positions, and connections between atoms. Thus, considering the three elements, an approach of Mamba-driven multi-perspective structural understanding (MPSU-Mamba) is proposed to localize molecular ground-state conformation. Particularly, for complex and diverse molecules, three different kinds of dedicated scanning strategies are explored to construct a comprehensive perception of corresponding molecular structures. And a bright-channel guided mechanism is defined to discriminate the critical conformation-related atom information. Experimental results on QM9 and Molecule3D datasets indicate that MPSU-Mamba significantly outperforms existing methods. Furthermore, we observe that for the case of few training samples, MPSU-Mamba still achieves superior performance, demonstrating that our method is indeed beneficial for understanding molecular structures.