A meta-learning approach for selectivity prediction in asymmetric catalysis
Abstract Transition metal-catalyzed asymmetric reactions are of high contemporary importance in organic synthesis. Recently, machine learning (ML) has shown promise in accelerating the development of newer catalytic protocols. However, the need for large amount of experimental data can present a bot...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-04-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-025-58854-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850181893330305024 |
|---|---|
| author | Sukriti Singh José Miguel Hernández-Lobato |
| author_facet | Sukriti Singh José Miguel Hernández-Lobato |
| author_sort | Sukriti Singh |
| collection | DOAJ |
| description | Abstract Transition metal-catalyzed asymmetric reactions are of high contemporary importance in organic synthesis. Recently, machine learning (ML) has shown promise in accelerating the development of newer catalytic protocols. However, the need for large amount of experimental data can present a bottleneck for implementing ML models. Here, we propose a meta-learning workflow that can harness the literature-derived data to extract shared reaction features and requires only a few examples to predict the outcome of new reactions. Prototypical networks are used as a meta-learning method to predict the enantioselectivity of asymmetric hydrogenation of olefins. This meta-learning model consistently provides significant performance improvement over other popular ML methods such as random forests and graph neural networks. The performance of our meta-model is analyzed with varying sizes of training examples to demonstrate its utility even with limited data. A good model performance on an out-of-sample test set further indicates the general applicability of our approach. We believe this work will provide a leap forward in identifying promising reactions in the early phases of reaction development when minimal data is available. |
| format | Article |
| id | doaj-art-6ce111a68d7d4769a3a5ae68d2a189b3 |
| institution | OA Journals |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-6ce111a68d7d4769a3a5ae68d2a189b32025-08-20T02:17:48ZengNature PortfolioNature Communications2041-17232025-04-0116111110.1038/s41467-025-58854-8A meta-learning approach for selectivity prediction in asymmetric catalysisSukriti Singh0José Miguel Hernández-Lobato1Department of Engineering, University of CambridgeDepartment of Engineering, University of CambridgeAbstract Transition metal-catalyzed asymmetric reactions are of high contemporary importance in organic synthesis. Recently, machine learning (ML) has shown promise in accelerating the development of newer catalytic protocols. However, the need for large amount of experimental data can present a bottleneck for implementing ML models. Here, we propose a meta-learning workflow that can harness the literature-derived data to extract shared reaction features and requires only a few examples to predict the outcome of new reactions. Prototypical networks are used as a meta-learning method to predict the enantioselectivity of asymmetric hydrogenation of olefins. This meta-learning model consistently provides significant performance improvement over other popular ML methods such as random forests and graph neural networks. The performance of our meta-model is analyzed with varying sizes of training examples to demonstrate its utility even with limited data. A good model performance on an out-of-sample test set further indicates the general applicability of our approach. We believe this work will provide a leap forward in identifying promising reactions in the early phases of reaction development when minimal data is available.https://doi.org/10.1038/s41467-025-58854-8 |
| spellingShingle | Sukriti Singh José Miguel Hernández-Lobato A meta-learning approach for selectivity prediction in asymmetric catalysis Nature Communications |
| title | A meta-learning approach for selectivity prediction in asymmetric catalysis |
| title_full | A meta-learning approach for selectivity prediction in asymmetric catalysis |
| title_fullStr | A meta-learning approach for selectivity prediction in asymmetric catalysis |
| title_full_unstemmed | A meta-learning approach for selectivity prediction in asymmetric catalysis |
| title_short | A meta-learning approach for selectivity prediction in asymmetric catalysis |
| title_sort | meta learning approach for selectivity prediction in asymmetric catalysis |
| url | https://doi.org/10.1038/s41467-025-58854-8 |
| work_keys_str_mv | AT sukritisingh ametalearningapproachforselectivitypredictioninasymmetriccatalysis AT josemiguelhernandezlobato ametalearningapproachforselectivitypredictioninasymmetriccatalysis AT sukritisingh metalearningapproachforselectivitypredictioninasymmetriccatalysis AT josemiguelhernandezlobato metalearningapproachforselectivitypredictioninasymmetriccatalysis |