Molecular ground-state conformation prediction based on the Mamba state space model
- 1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China
Abstract
A comprehensive understanding of molecular structures is important for the prediction of molecular ground-state conformation involving property information. Recently, state space models (e.g., Mamba) have emerged as a powerful paradigm for sequence modeling, and have achieved remarkable success in 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 effective 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 an Max-Activation Gating 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.
Related
- Mamba Quantum Eigensolver (MQE): A Selective State Space Model for Scalable Ground State Simulations in Quantum Chemistry
- Literature-Scale Mössbauer Data Enable Prediction, Electronic-State Interpretation, and Chemical Insight in Molecular Iron Chemistry
- Molecular Conformation as a Determinant of Self-Assembly in Sequence-Isomeric Tripeptides.
- Phys-Mamba: Physics-informed selective state-space fusion network for high-fidelity underwater image restoration
- Conformational Soft Hash (CSH): A Rotation-Invariant Molecular Fingerprint for Conformation-Sensitive Similarity Search and Blockchain-Based Conformational Registry
BibTeX
@article{gou2026molecular,
title = {{Molecular ground-state conformation prediction based on the Mamba state space model}},
author = {Yuxin Gou and Aming Wu and Richang Hong and Meng Wang},
journal = {Communications Chemistry},
year = {2026},
month = jul,
doi = {10.1038/s42004-026-02113-4},
url = {https://doi.org/10.1038/s42004-026-02113-4},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Molecular ground-state conformation prediction based on the Mamba state space model
AU - Yuxin Gou
AU - Aming Wu
AU - Richang Hong
AU - Meng Wang
PY - 2026
DA - 2026/07/06
JO - Communications Chemistry
DO - 10.1038/s42004-026-02113-4
UR - https://doi.org/10.1038/s42004-026-02113-4
AB - A comprehensive understanding of molecular structures is important for the prediction of molecular ground-state conformation involving property information. Recently, state space models (e.g., Mamba) have emerged as a powerful paradigm for sequence modeling, and have achieved remarkable success in 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 effective 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 an Max-Activation Gating 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.
SN - 2399-3669
ER -