Chem3DLLM: 3D Multimodal Large Language Models for Chemistry
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916899334389760 |
|---|---|
| author | Jiang, Lei Sun, Shuzhou Qi, Biqing Fu, Yuchen Xu, Xiaohua Li, Yuqiang Zhou, Dongzhan Fu, Tianfan |
| author_facet | Jiang, Lei Sun, Shuzhou Qi, Biqing Fu, Yuchen Xu, Xiaohua Li, Yuqiang Zhou, Dongzhan Fu, Tianfan |
| contents | In the real world, a molecule is a 3D geometric structure. Compared to 1D SMILES sequences and 2D molecular graphs, 3D molecules represent the most informative molecular modality. Despite the rapid progress of autoregressive-based language models, they cannot handle the generation of 3D molecular conformation due to several challenges: 1) 3D molecular structures are incompatible with LLMs' discrete token space, 2) integrating heterogeneous inputs like proteins, ligands, and text remains difficult within a unified model, and 3) LLMs lack essential scientific priors, hindering the enforcement of physical and chemical constraints during generation. To tackle these issues, we present Chem3DLLM, a unified protein-conditioned multimodal large language model. Our approach designs a novel reversible text encoding for 3D molecular structures using run-length compression, achieving 3x size reduction while preserving complete structural information. This enables seamless integration of molecular geometry with protein pocket features in a single LLM architecture. We employ reinforcement learning with stability-based rewards to optimize chemical validity and incorporate a lightweight protein embedding projector for end-to-end training. Experimental results on structure-based drug design demonstrate state-of-the-art performance with a Vina score of -7.21, validating our unified multimodal approach for practical drug discovery applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_10696 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Chem3DLLM: 3D Multimodal Large Language Models for Chemistry Jiang, Lei Sun, Shuzhou Qi, Biqing Fu, Yuchen Xu, Xiaohua Li, Yuqiang Zhou, Dongzhan Fu, Tianfan Computational Engineering, Finance, and Science In the real world, a molecule is a 3D geometric structure. Compared to 1D SMILES sequences and 2D molecular graphs, 3D molecules represent the most informative molecular modality. Despite the rapid progress of autoregressive-based language models, they cannot handle the generation of 3D molecular conformation due to several challenges: 1) 3D molecular structures are incompatible with LLMs' discrete token space, 2) integrating heterogeneous inputs like proteins, ligands, and text remains difficult within a unified model, and 3) LLMs lack essential scientific priors, hindering the enforcement of physical and chemical constraints during generation. To tackle these issues, we present Chem3DLLM, a unified protein-conditioned multimodal large language model. Our approach designs a novel reversible text encoding for 3D molecular structures using run-length compression, achieving 3x size reduction while preserving complete structural information. This enables seamless integration of molecular geometry with protein pocket features in a single LLM architecture. We employ reinforcement learning with stability-based rewards to optimize chemical validity and incorporate a lightweight protein embedding projector for end-to-end training. Experimental results on structure-based drug design demonstrate state-of-the-art performance with a Vina score of -7.21, validating our unified multimodal approach for practical drug discovery applications. |
| title | Chem3DLLM: 3D Multimodal Large Language Models for Chemistry |
| topic | Computational Engineering, Finance, and Science |
| url | https://arxiv.org/abs/2508.10696 |