Equivariant score-based generative diffusion framework for 3D molecules
Abstract Background Molecular biology is crucial for drug discovery, protein design, and human health. Due to the vastness of the drug-like chemical space, depending on biomedical experts to manually design molecules is exceedingly expensive. Utilizing generative methods with deep learning technolog...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2024-05-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12859-024-05810-w |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846164835643949056 |
|---|---|
| author | Hao Zhang Yang Liu Xiaoyan Liu Cheng Wang Maozu Guo |
| author_facet | Hao Zhang Yang Liu Xiaoyan Liu Cheng Wang Maozu Guo |
| author_sort | Hao Zhang |
| collection | DOAJ |
| description | Abstract Background Molecular biology is crucial for drug discovery, protein design, and human health. Due to the vastness of the drug-like chemical space, depending on biomedical experts to manually design molecules is exceedingly expensive. Utilizing generative methods with deep learning technology offers an effective approach to streamline the search space for molecular design and save costs. This paper introduces a novel E(3)-equivariant score-based diffusion framework for 3D molecular generation via SDEs, aiming to address the constraints of unified Gaussian diffusion methods. Within the proposed framework EMDS, the complete diffusion is decomposed into separate diffusion processes for distinct components of the molecular feature space, while the modeling processes also capture the complex dependency among these components. Moreover, angle and torsion angle information is integrated into the networks to enhance the modeling of atom coordinates and utilize spatial information more effectively. Results Experiments on the widely utilized QM9 dataset demonstrate that our proposed framework significantly outperforms the state-of-the-art methods in all evaluation metrics for 3D molecular generation. Additionally, ablation experiments are conducted to highlight the contribution of key components in our framework, demonstrating the effectiveness of the proposed framework and the performance improvements of incorporating angle and torsion angle information for molecular generation. Finally, the comparative results of distribution show that our method is highly effective in generating molecules that closely resemble the actual scenario. Conclusion Through the experiments and comparative results, our framework clearly outperforms previous 3D molecular generation methods, exhibiting significantly better capacity for modeling chemically realistic molecules. The excellent performance of EMDS in 3D molecular generation brings novel and encouraging opportunities for tackling challenging biomedical molecule and protein scenarios. |
| format | Article |
| id | doaj-art-bae72ea5a1a5458ca683b6e2fa0e2b8c |
| institution | Kabale University |
| issn | 1471-2105 |
| language | English |
| publishDate | 2024-05-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj-art-bae72ea5a1a5458ca683b6e2fa0e2b8c2024-11-17T12:51:19ZengBMCBMC Bioinformatics1471-21052024-05-0125111910.1186/s12859-024-05810-wEquivariant score-based generative diffusion framework for 3D moleculesHao Zhang0Yang Liu1Xiaoyan Liu2Cheng Wang3Maozu Guo4School of Computer Science and Technology, Harbin Institute of TechnologySchool of Computer Science and Technology, Harbin Institute of TechnologySchool of Computer Science and Technology, Harbin Institute of TechnologySchool of Computer Science and Technology, Harbin Institute of TechnologySchool of Electrical and Information Engineering, Beijing University of Civil Engineering and ArchitectureAbstract Background Molecular biology is crucial for drug discovery, protein design, and human health. Due to the vastness of the drug-like chemical space, depending on biomedical experts to manually design molecules is exceedingly expensive. Utilizing generative methods with deep learning technology offers an effective approach to streamline the search space for molecular design and save costs. This paper introduces a novel E(3)-equivariant score-based diffusion framework for 3D molecular generation via SDEs, aiming to address the constraints of unified Gaussian diffusion methods. Within the proposed framework EMDS, the complete diffusion is decomposed into separate diffusion processes for distinct components of the molecular feature space, while the modeling processes also capture the complex dependency among these components. Moreover, angle and torsion angle information is integrated into the networks to enhance the modeling of atom coordinates and utilize spatial information more effectively. Results Experiments on the widely utilized QM9 dataset demonstrate that our proposed framework significantly outperforms the state-of-the-art methods in all evaluation metrics for 3D molecular generation. Additionally, ablation experiments are conducted to highlight the contribution of key components in our framework, demonstrating the effectiveness of the proposed framework and the performance improvements of incorporating angle and torsion angle information for molecular generation. Finally, the comparative results of distribution show that our method is highly effective in generating molecules that closely resemble the actual scenario. Conclusion Through the experiments and comparative results, our framework clearly outperforms previous 3D molecular generation methods, exhibiting significantly better capacity for modeling chemically realistic molecules. The excellent performance of EMDS in 3D molecular generation brings novel and encouraging opportunities for tackling challenging biomedical molecule and protein scenarios.https://doi.org/10.1186/s12859-024-05810-w3D molecular generationScore-based diffusion modelPartial score functionsE(3)-equivariant |
| spellingShingle | Hao Zhang Yang Liu Xiaoyan Liu Cheng Wang Maozu Guo Equivariant score-based generative diffusion framework for 3D molecules BMC Bioinformatics 3D molecular generation Score-based diffusion model Partial score functions E(3)-equivariant |
| title | Equivariant score-based generative diffusion framework for 3D molecules |
| title_full | Equivariant score-based generative diffusion framework for 3D molecules |
| title_fullStr | Equivariant score-based generative diffusion framework for 3D molecules |
| title_full_unstemmed | Equivariant score-based generative diffusion framework for 3D molecules |
| title_short | Equivariant score-based generative diffusion framework for 3D molecules |
| title_sort | equivariant score based generative diffusion framework for 3d molecules |
| topic | 3D molecular generation Score-based diffusion model Partial score functions E(3)-equivariant |
| url | https://doi.org/10.1186/s12859-024-05810-w |
| work_keys_str_mv | AT haozhang equivariantscorebasedgenerativediffusionframeworkfor3dmolecules AT yangliu equivariantscorebasedgenerativediffusionframeworkfor3dmolecules AT xiaoyanliu equivariantscorebasedgenerativediffusionframeworkfor3dmolecules AT chengwang equivariantscorebasedgenerativediffusionframeworkfor3dmolecules AT maozuguo equivariantscorebasedgenerativediffusionframeworkfor3dmolecules |