Unlocking comprehensive molecular design across all scenarios with large language model and unordered chemical language-Reference-Cited by-同舟云学术

Unlocking comprehensive molecular design across all scenarios with large language model and unordered chemical language

Published:2024 Issue:34 Volume:15 Page:13727-13740
ISSN:2041-6520
Container-title:Chemical Science
language:en
Short-container-title:Chem. Sci.

Author:

Yue Jie¹,Peng Bingxin¹²,Chen Yu²,Jin Jieyu³,Zhao Xinda²,Shen Chao³²^ORCID,Ji Xiangyang⁴,Hsieh Chang-Yu³²^ORCID,Song Jianfei²,Hou Tingjun³²^ORCID,Deng Yafeng²⁴,Wang Jike³²

Affiliation:

1. College of Information Engineering, Hebei University of Architecture, Zhangjiakou 075132, Hebei, China

2. CarbonSilicon AI Technology Co., Ltd, Hangzhou 310018, Zhejiang, China

3. College of Pharmaceutical Sciences, Zhejiang University, Hangzhou 310058, Zhejiang, China

4. Department of Automation, Tsinghua University, Beijing 100084, China

Abstract

We introduce FU-SMILES and a versatile fragment-based molecular generation model, FragGPT. Leveraging FU-SMILES, FragGPT enables efficient molecular generation across a range of practical applications.

Funder

National Key Research and Development Program of China

National Natural Science Foundation of China

China Postdoctoral Science Foundation

Publisher

Royal Society of Chemistry (RSC)

Link

http://pubs.rsc.org/en/content/articlepdf/2024/SC/D4SC03744H

Reference61 articles.

1. SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules

2. MolGPT: Molecular Generation Using a Transformer-Decoder Model

3. J.-N.Wu , T.Wang , Y.Chen , L.-J.Tang , H.-L.Wu and R.-Q.Yu , arXiv , 2023 , preprint, arXiv:2301.01829, 10.48550/arXiv.2301.01829

4. Multi-constraint molecular generation based on conditional transformer, knowledge distillation and reinforcement learning

5. REINVENT 2.0: An AI Tool for De Novo Drug Design