BindGPT: A Scalable Framework for 3D Molecular Design via Language Modeling and Reinforcement Learning

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zholus, Artem, Kuznetsov, Maksim, Schutski, Roman, Shayakhmetov, Rim, Polykovskiy, Daniil, Chandar, Sarath, Zhavoronkov, Alex
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917686385049600
author Zholus, Artem
Kuznetsov, Maksim
Schutski, Roman
Shayakhmetov, Rim
Polykovskiy, Daniil
Chandar, Sarath
Zhavoronkov, Alex
author_facet Zholus, Artem
Kuznetsov, Maksim
Schutski, Roman
Shayakhmetov, Rim
Polykovskiy, Daniil
Chandar, Sarath
Zhavoronkov, Alex
contents Generating novel active molecules for a given protein is an extremely challenging task for generative models that requires an understanding of the complex physical interactions between the molecule and its environment. In this paper, we present a novel generative model, BindGPT which uses a conceptually simple but powerful approach to create 3D molecules within the protein's binding site. Our model produces molecular graphs and conformations jointly, eliminating the need for an extra graph reconstruction step. We pretrain BindGPT on a large-scale dataset and fine-tune it with reinforcement learning using scores from external simulation software. We demonstrate how a single pretrained language model can serve at the same time as a 3D molecular generative model, conformer generator conditioned on the molecular graph, and a pocket-conditioned 3D molecule generator. Notably, the model does not make any representational equivariance assumptions about the domain of generation. We show how such simple conceptual approach combined with pretraining and scaling can perform on par or better than the current best specialized diffusion models, language models, and graph neural networks while being two orders of magnitude cheaper to sample.
format Preprint
id arxiv_https___arxiv_org_abs_2406_03686
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle BindGPT: A Scalable Framework for 3D Molecular Design via Language Modeling and Reinforcement Learning
Zholus, Artem
Kuznetsov, Maksim
Schutski, Roman
Shayakhmetov, Rim
Polykovskiy, Daniil
Chandar, Sarath
Zhavoronkov, Alex
Machine Learning
Generating novel active molecules for a given protein is an extremely challenging task for generative models that requires an understanding of the complex physical interactions between the molecule and its environment. In this paper, we present a novel generative model, BindGPT which uses a conceptually simple but powerful approach to create 3D molecules within the protein's binding site. Our model produces molecular graphs and conformations jointly, eliminating the need for an extra graph reconstruction step. We pretrain BindGPT on a large-scale dataset and fine-tune it with reinforcement learning using scores from external simulation software. We demonstrate how a single pretrained language model can serve at the same time as a 3D molecular generative model, conformer generator conditioned on the molecular graph, and a pocket-conditioned 3D molecule generator. Notably, the model does not make any representational equivariance assumptions about the domain of generation. We show how such simple conceptual approach combined with pretraining and scaling can perform on par or better than the current best specialized diffusion models, language models, and graph neural networks while being two orders of magnitude cheaper to sample.
title BindGPT: A Scalable Framework for 3D Molecular Design via Language Modeling and Reinforcement Learning
topic Machine Learning
url https://arxiv.org/abs/2406.03686