G 2 Retro as a two-step graph generative models for retrosynthesis prediction
Abstract Retrosynthesis is a procedure where a target molecule is transformed into potential reactants and thus the synthesis routes can be identified. Recently, computational approaches have been developed to accelerate the design of synthesis routes. In this paper,we develop a generative framework...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-05-01
|
| Series: | Communications Chemistry |
| Online Access: | https://doi.org/10.1038/s42004-023-00897-3 |
| _version_ | 1797811609279660032 |
|---|---|
| author | Ziqi Chen Oluwatosin R. Ayinde James R. Fuchs Huan Sun Xia Ning |
| author_facet | Ziqi Chen Oluwatosin R. Ayinde James R. Fuchs Huan Sun Xia Ning |
| author_sort | Ziqi Chen |
| collection | DOAJ |
| description | Abstract Retrosynthesis is a procedure where a target molecule is transformed into potential reactants and thus the synthesis routes can be identified. Recently, computational approaches have been developed to accelerate the design of synthesis routes. In this paper,we develop a generative framework G 2 Retro for one-step retrosynthesis prediction. G 2 Retro imitates the reversed logic of synthetic reactions. It first predicts the reaction centers in the target molecules (products), identifies the synthons needed to assemble the products, and transforms these synthons into reactants. G 2 Retro defines a comprehensive set of reaction center types, and learns from the molecular graphs of the products to predict potential reaction centers. To complete synthons into reactants, G 2 Retro considers all the involved synthon structures and the product structures to identify the optimal completion paths, and accordingly attaches small substructures sequentially to the synthons. Here we show that G 2 Retro is able to better predict the reactants for given products in the benchmark dataset than the state-of-the-art methods. |
| first_indexed | 2024-03-13T07:25:13Z |
| format | Article |
| id | doaj.art-480fb60db89f4f459d39885a7d8ea005 |
| institution | Directory Open Access Journal |
| issn | 2399-3669 |
| language | English |
| last_indexed | 2024-03-13T07:25:13Z |
| publishDate | 2023-05-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Chemistry |
| spelling | doaj.art-480fb60db89f4f459d39885a7d8ea0052023-06-04T11:23:13ZengNature PortfolioCommunications Chemistry2399-36692023-05-016111910.1038/s42004-023-00897-3G 2 Retro as a two-step graph generative models for retrosynthesis predictionZiqi Chen0Oluwatosin R. Ayinde1James R. Fuchs2Huan Sun3Xia Ning4Computer Science and Engineering, The Ohio State UniversityMedicinal Chemistry and Pharmacognosy, College of Pharmacy, The Ohio State UniversityMedicinal Chemistry and Pharmacognosy, College of Pharmacy, The Ohio State UniversityComputer Science and Engineering, The Ohio State UniversityComputer Science and Engineering, The Ohio State UniversityAbstract Retrosynthesis is a procedure where a target molecule is transformed into potential reactants and thus the synthesis routes can be identified. Recently, computational approaches have been developed to accelerate the design of synthesis routes. In this paper,we develop a generative framework G 2 Retro for one-step retrosynthesis prediction. G 2 Retro imitates the reversed logic of synthetic reactions. It first predicts the reaction centers in the target molecules (products), identifies the synthons needed to assemble the products, and transforms these synthons into reactants. G 2 Retro defines a comprehensive set of reaction center types, and learns from the molecular graphs of the products to predict potential reaction centers. To complete synthons into reactants, G 2 Retro considers all the involved synthon structures and the product structures to identify the optimal completion paths, and accordingly attaches small substructures sequentially to the synthons. Here we show that G 2 Retro is able to better predict the reactants for given products in the benchmark dataset than the state-of-the-art methods.https://doi.org/10.1038/s42004-023-00897-3 |
| spellingShingle | Ziqi Chen Oluwatosin R. Ayinde James R. Fuchs Huan Sun Xia Ning G 2 Retro as a two-step graph generative models for retrosynthesis prediction Communications Chemistry |
| title | G 2 Retro as a two-step graph generative models for retrosynthesis prediction |
| title_full | G 2 Retro as a two-step graph generative models for retrosynthesis prediction |
| title_fullStr | G 2 Retro as a two-step graph generative models for retrosynthesis prediction |
| title_full_unstemmed | G 2 Retro as a two-step graph generative models for retrosynthesis prediction |
| title_short | G 2 Retro as a two-step graph generative models for retrosynthesis prediction |
| title_sort | g 2 retro as a two step graph generative models for retrosynthesis prediction |
| url | https://doi.org/10.1038/s42004-023-00897-3 |
| work_keys_str_mv | AT ziqichen g2retroasatwostepgraphgenerativemodelsforretrosynthesisprediction AT oluwatosinrayinde g2retroasatwostepgraphgenerativemodelsforretrosynthesisprediction AT jamesrfuchs g2retroasatwostepgraphgenerativemodelsforretrosynthesisprediction AT huansun g2retroasatwostepgraphgenerativemodelsforretrosynthesisprediction AT xianing g2retroasatwostepgraphgenerativemodelsforretrosynthesisprediction |