Pre-training of Molecular GNNs via Conditional Boltzmann Generator
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910302409326592 |
|---|---|
| author | Koge, Daiki Ono, Naoaki Kanaya, Shigehiko |
| author_facet | Koge, Daiki Ono, Naoaki Kanaya, Shigehiko |
| contents | Learning representations of molecular structures using deep learning is a fundamental problem in molecular property prediction tasks. Molecules inherently exist in the real world as three-dimensional structures; furthermore, they are not static but in continuous motion in the 3D Euclidean space, forming a potential energy surface. Therefore, it is desirable to generate multiple conformations in advance and extract molecular representations using a 4D-QSAR model that incorporates multiple conformations. However, this approach is impractical for drug and material discovery tasks because of the computational cost of obtaining multiple conformations. To address this issue, we propose a pre-training method for molecular GNNs using an existing dataset of molecular conformations to generate a latent vector universal to multiple conformations from a 2D molecular graph. Our method, called Boltzmann GNN, is formulated by maximizing the conditional marginal likelihood of a conditional generative model for conformations generation. We show that our model has a better prediction performance for molecular properties than existing pre-training methods using molecular graphs and three-dimensional molecular structures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_13110 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Pre-training of Molecular GNNs via Conditional Boltzmann Generator Koge, Daiki Ono, Naoaki Kanaya, Shigehiko Machine Learning Chemical Physics Biomolecules Learning representations of molecular structures using deep learning is a fundamental problem in molecular property prediction tasks. Molecules inherently exist in the real world as three-dimensional structures; furthermore, they are not static but in continuous motion in the 3D Euclidean space, forming a potential energy surface. Therefore, it is desirable to generate multiple conformations in advance and extract molecular representations using a 4D-QSAR model that incorporates multiple conformations. However, this approach is impractical for drug and material discovery tasks because of the computational cost of obtaining multiple conformations. To address this issue, we propose a pre-training method for molecular GNNs using an existing dataset of molecular conformations to generate a latent vector universal to multiple conformations from a 2D molecular graph. Our method, called Boltzmann GNN, is formulated by maximizing the conditional marginal likelihood of a conditional generative model for conformations generation. We show that our model has a better prediction performance for molecular properties than existing pre-training methods using molecular graphs and three-dimensional molecular structures. |
| title | Pre-training of Molecular GNNs via Conditional Boltzmann Generator |
| topic | Machine Learning Chemical Physics Biomolecules |
| url | https://arxiv.org/abs/2312.13110 |