GraphXForm: Graph transformer for computer-aided molecular design
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913746710953984 |
|---|---|
| author | Pirnay, Jonathan Rittig, Jan G. Wolf, Alexander B. Grohe, Martin Burger, Jakob Mitsos, Alexander Grimm, Dominik G. |
| author_facet | Pirnay, Jonathan Rittig, Jan G. Wolf, Alexander B. Grohe, Martin Burger, Jakob Mitsos, Alexander Grimm, Dominik G. |
| contents | Generative deep learning has become pivotal in molecular design for drug discovery, materials science, and chemical engineering. A widely used paradigm is to pretrain neural networks on string representations of molecules and fine-tune them using reinforcement learning on specific objectives. However, string-based models face challenges in ensuring chemical validity and enforcing structural constraints like the presence of specific substructures. We propose to instead combine graph-based molecular representations, which can naturally ensure chemical validity, with transformer architectures, which are highly expressive and capable of modeling long-range dependencies between atoms. Our approach iteratively modifies a molecular graph by adding atoms and bonds, which ensures chemical validity and facilitates the incorporation of structural constraints. We present GraphXForm, a decoder-only graph transformer architecture, which is pretrained on existing compounds and then fine-tuned using a new training algorithm that combines elements of the deep cross-entropy method and self-improvement learning. We evaluate GraphXForm on various drug design tasks, demonstrating superior objective scores compared to state-of-the-art molecular design approaches. Furthermore, we apply GraphXForm to two solvent design tasks for liquid-liquid extraction, again outperforming alternative methods while flexibly enforcing structural constraints or initiating design from existing molecular structures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_01667 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GraphXForm: Graph transformer for computer-aided molecular design Pirnay, Jonathan Rittig, Jan G. Wolf, Alexander B. Grohe, Martin Burger, Jakob Mitsos, Alexander Grimm, Dominik G. Machine Learning Chemical Physics Biomolecules Generative deep learning has become pivotal in molecular design for drug discovery, materials science, and chemical engineering. A widely used paradigm is to pretrain neural networks on string representations of molecules and fine-tune them using reinforcement learning on specific objectives. However, string-based models face challenges in ensuring chemical validity and enforcing structural constraints like the presence of specific substructures. We propose to instead combine graph-based molecular representations, which can naturally ensure chemical validity, with transformer architectures, which are highly expressive and capable of modeling long-range dependencies between atoms. Our approach iteratively modifies a molecular graph by adding atoms and bonds, which ensures chemical validity and facilitates the incorporation of structural constraints. We present GraphXForm, a decoder-only graph transformer architecture, which is pretrained on existing compounds and then fine-tuned using a new training algorithm that combines elements of the deep cross-entropy method and self-improvement learning. We evaluate GraphXForm on various drug design tasks, demonstrating superior objective scores compared to state-of-the-art molecular design approaches. Furthermore, we apply GraphXForm to two solvent design tasks for liquid-liquid extraction, again outperforming alternative methods while flexibly enforcing structural constraints or initiating design from existing molecular structures. |
| title | GraphXForm: Graph transformer for computer-aided molecular design |
| topic | Machine Learning Chemical Physics Biomolecules |
| url | https://arxiv.org/abs/2411.01667 |