1
|
Wang Y, Pang C, Wang Y, Jin J, Zhang J, Zeng X, Su R, Zou Q, Wei L. Retrosynthesis prediction with an interpretable deep-learning framework based on molecular assembly tasks. Nat Commun 2023; 14:6155. [PMID: 37788995 PMCID: PMC10547708 DOI: 10.1038/s41467-023-41698-5] [Citation(s) in RCA: 11] [Impact Index Per Article: 11.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/16/2023] [Accepted: 09/12/2023] [Indexed: 10/05/2023] Open
Abstract
Automating retrosynthesis with artificial intelligence expedites organic chemistry research in digital laboratories. However, most existing deep-learning approaches are hard to explain, like a "black box" with few insights. Here, we propose RetroExplainer, formulizing the retrosynthesis task into a molecular assembly process, containing several retrosynthetic actions guided by deep learning. To guarantee a robust performance of our model, we propose three units: a multi-sense and multi-scale Graph Transformer, structure-aware contrastive learning, and dynamic adaptive multi-task learning. The results on 12 large-scale benchmark datasets demonstrate the effectiveness of RetroExplainer, which outperforms the state-of-the-art single-step retrosynthesis approaches. In addition, the molecular assembly process renders our model with good interpretability, allowing for transparent decision-making and quantitative attribution. When extended to multi-step retrosynthesis planning, RetroExplainer has identified 101 pathways, in which 86.9% of the single reactions correspond to those already reported in the literature. As a result, RetroExplainer is expected to offer valuable insights for reliable, high-throughput, and high-quality organic synthesis in drug development.
Collapse
Affiliation(s)
- Yu Wang
- School of Software, Shandong University, Jinan, 250101, China
- Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR), Shandong University, Jinan, 250101, China
| | - Chao Pang
- School of Software, Shandong University, Jinan, 250101, China
- Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR), Shandong University, Jinan, 250101, China
| | - Yuzhe Wang
- School of Software, Shandong University, Jinan, 250101, China
- Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR), Shandong University, Jinan, 250101, China
| | - Junru Jin
- School of Software, Shandong University, Jinan, 250101, China
- Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR), Shandong University, Jinan, 250101, China
| | - Jingjie Zhang
- School of Software, Shandong University, Jinan, 250101, China
- Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR), Shandong University, Jinan, 250101, China
| | - Xiangxiang Zeng
- College of Computer Science and Electronic Engineering, Hunan University, Changsha, China
| | - Ran Su
- College of Intelligence and Computing, Tianjin University, Tianjin, China
| | - Quan Zou
- Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu, China.
| | - Leyi Wei
- School of Software, Shandong University, Jinan, 250101, China.
- College of Computer Science and Electronic Engineering, Hunan University, Changsha, China.
| |
Collapse
|
2
|
Abstract
Ecstasy is a widely used recreational drug that usually consists primarily of 3,4-methylenedioxymethamphetamine (MDMA). Most ecstasy users consume other substances as well, which complicates the interpretation of research in this field. The positively rated effects of MDMA consumption include euphoria, arousal, enhanced mood, increased sociability, and heightened perceptions; some common adverse reactions are nausea, headache, tachycardia, bruxism, and trismus. Lowering of mood is an aftereffect that is sometimes reported from 2 to 5 days after a session of ecstasy use. The acute effects of MDMA in ecstasy users have been attributed primarily to increased release and inhibited reuptake of serotonin (5-HT) and norepinephrine, along with possible release of the neuropeptide oxytocin. Repeated or high-dose MDMA/ecstasy use has been associated with tolerance, depressive symptomatology, and persisting cognitive deficits, particularly in memory tests. Animal studies have demonstrated that high doses of MDMA can lead to long-term decreases in forebrain 5-HT concentrations, tryptophan hydroxylase activity, serotonin transporter (SERT) expression, and visualization of axons immunoreactive for 5-HT or SERT. These neurotoxic effects may reflect either a drug-induced degeneration of serotonergic fibers or a long-lasting downregulation in 5-HT and SERT biosynthesis. Possible neurotoxicity in heavy ecstasy users has been revealed by neuroimaging studies showing reduced SERT binding and increased 5-HT2A receptor binding in several cortical and/or subcortical areas. MDMA overdose or use with certain other drugs can also cause severe morbidity and even death. Repeated use of MDMA may lead to dose escalation and the development of dependence, although such dependence is usually not as profound as is seen with many other drugs of abuse. MDMA/ecstasy-dependent patients are treated with standard addiction programs, since there are no specific programs for this substance and no proven medications. Finally, even though MDMA is listed as a Schedule I compound by the Drug Enforcement Agency, MDMA-assisted psychotherapy for patients with chronic, treatment-resistant posttraumatic stress disorder is currently under investigation. Initial results show efficacy for this treatment approach, although considerably more research must be performed to confirm such efficacy and to ensure that the benefits of MDMA-assisted therapy outweigh the risks to the patients.
Collapse
Affiliation(s)
- Jerrold S Meyer
- Department of Psychology, Neuroscience and Behavior Program, University of Massachusetts, Amherst, MA, USA
| |
Collapse
|
3
|
Martinez CM, Neudörffer A, Largeron M. A convenient biomimetic synthesis of optically active putative neurotoxic metabolites of MDMA (“ecstasy”) from R-(−)- and S-(+)-N-methyl-α-methyldopamine precursors. Org Biomol Chem 2012; 10:3739-48. [DOI: 10.1039/c2ob25245g] [Citation(s) in RCA: 6] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/21/2022]
|