Precise atom-to-atom mapping for organic reactions via human-in-the-loop machine learning
Abstract Atom-to-atom mapping (AAM) is a task of identifying the position of each atom in the molecules before and after a chemical reaction, which is important for understanding the reaction mechanism. As more machine learning (ML) models were developed for retrosynthesis and reaction outcome predi...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-03-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-024-46364-y |
| _version_ | 1797259130033930240 |
|---|---|
| author | Shuan Chen Sunggi An Ramil Babazade Yousung Jung |
| author_facet | Shuan Chen Sunggi An Ramil Babazade Yousung Jung |
| author_sort | Shuan Chen |
| collection | DOAJ |
| description | Abstract Atom-to-atom mapping (AAM) is a task of identifying the position of each atom in the molecules before and after a chemical reaction, which is important for understanding the reaction mechanism. As more machine learning (ML) models were developed for retrosynthesis and reaction outcome prediction recently, the quality of these models is highly dependent on the quality of the AAM in reaction datasets. Although there are algorithms using graph theory or unsupervised learning to label the AAM for reaction datasets, existing methods map the atoms based on substructure alignments instead of chemistry knowledge. Here, we present LocalMapper, an ML model that learns correct AAM from chemist-labeled reactions via human-in-the-loop machine learning. We show that LocalMapper can predict the AAM for 50 K reactions with 98.5% calibrated accuracy by learning from only 2% of the human-labeled reactions from the entire dataset. More importantly, the confident predictions given by LocalMapper, which cover 97% of 50 K reactions, show 100% accuracy for 3,000 randomly sampled reactions. In an out-of-distribution experiment, LocalMapper shows favorable performance over other existing methods. We expect LocalMapper can be used to generate more precise reaction AAM and improve the quality of future ML-based reaction prediction models. |
| first_indexed | 2024-04-24T23:04:32Z |
| format | Article |
| id | doaj.art-28154dc9a99c4da49aed3084dd3a8fe4 |
| institution | Directory Open Access Journal |
| issn | 2041-1723 |
| language | English |
| last_indexed | 2024-04-24T23:04:32Z |
| publishDate | 2024-03-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj.art-28154dc9a99c4da49aed3084dd3a8fe42024-03-17T12:31:17ZengNature PortfolioNature Communications2041-17232024-03-0115111010.1038/s41467-024-46364-yPrecise atom-to-atom mapping for organic reactions via human-in-the-loop machine learningShuan Chen0Sunggi An1Ramil Babazade2Yousung Jung3Department of Chemical and Biomolecular Engineering, KAISTDepartment of Chemical and Biomolecular Engineering, KAISTGraduate School of AI, KAISTDepartment of Chemical and Biomolecular Engineering, KAISTAbstract Atom-to-atom mapping (AAM) is a task of identifying the position of each atom in the molecules before and after a chemical reaction, which is important for understanding the reaction mechanism. As more machine learning (ML) models were developed for retrosynthesis and reaction outcome prediction recently, the quality of these models is highly dependent on the quality of the AAM in reaction datasets. Although there are algorithms using graph theory or unsupervised learning to label the AAM for reaction datasets, existing methods map the atoms based on substructure alignments instead of chemistry knowledge. Here, we present LocalMapper, an ML model that learns correct AAM from chemist-labeled reactions via human-in-the-loop machine learning. We show that LocalMapper can predict the AAM for 50 K reactions with 98.5% calibrated accuracy by learning from only 2% of the human-labeled reactions from the entire dataset. More importantly, the confident predictions given by LocalMapper, which cover 97% of 50 K reactions, show 100% accuracy for 3,000 randomly sampled reactions. In an out-of-distribution experiment, LocalMapper shows favorable performance over other existing methods. We expect LocalMapper can be used to generate more precise reaction AAM and improve the quality of future ML-based reaction prediction models.https://doi.org/10.1038/s41467-024-46364-y |
| spellingShingle | Shuan Chen Sunggi An Ramil Babazade Yousung Jung Precise atom-to-atom mapping for organic reactions via human-in-the-loop machine learning Nature Communications |
| title | Precise atom-to-atom mapping for organic reactions via human-in-the-loop machine learning |
| title_full | Precise atom-to-atom mapping for organic reactions via human-in-the-loop machine learning |
| title_fullStr | Precise atom-to-atom mapping for organic reactions via human-in-the-loop machine learning |
| title_full_unstemmed | Precise atom-to-atom mapping for organic reactions via human-in-the-loop machine learning |
| title_short | Precise atom-to-atom mapping for organic reactions via human-in-the-loop machine learning |
| title_sort | precise atom to atom mapping for organic reactions via human in the loop machine learning |
| url | https://doi.org/10.1038/s41467-024-46364-y |
| work_keys_str_mv | AT shuanchen preciseatomtoatommappingfororganicreactionsviahumanintheloopmachinelearning AT sunggian preciseatomtoatommappingfororganicreactionsviahumanintheloopmachinelearning AT ramilbabazade preciseatomtoatommappingfororganicreactionsviahumanintheloopmachinelearning AT yousungjung preciseatomtoatommappingfororganicreactionsviahumanintheloopmachinelearning |