Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines
Data-driven reaction discovery and development is a growing field that relies on the use of molecular descriptors to capture key information about substrates, ligands, and targets. Broad adaptation of this strategy is hindered by the associated computational cost of descriptor calculation, especiall...
| Main Authors: | , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Royal Society of Chemistry
2025
|
| Online Access: | https://hdl.handle.net/1721.1/158096 |
| _version_ | 1824458233494896640 |
|---|---|
| author | Haas, Brittany C Hardy, Melissa A Sowndarya S. V., Shree Adams, Keir Coley, Connor W Paton, Robert S Sigman, Matthew S |
| author2 | Massachusetts Institute of Technology. Department of Chemical Engineering |
| author_facet | Massachusetts Institute of Technology. Department of Chemical Engineering Haas, Brittany C Hardy, Melissa A Sowndarya S. V., Shree Adams, Keir Coley, Connor W Paton, Robert S Sigman, Matthew S |
| author_sort | Haas, Brittany C |
| collection | MIT |
| description | Data-driven reaction discovery and development is a growing field that relies on the use of molecular descriptors to capture key information about substrates, ligands, and targets. Broad adaptation of this strategy is hindered by the associated computational cost of descriptor calculation, especially when considering conformational flexibility. Descriptor libraries can be precomputed agnostic of application to reduce the computational burden of data-driven reaction development. However, as one often applies these models to evaluate novel hypothetical structures, it would be ideal to predict the descriptors of compounds on-the-fly. Herein, we report DFT-level descriptor libraries for conformational ensembles of 8528 carboxylic acids and 8172 alkyl amines towards this goal. Employing 2D and 3D graph neural network architectures trained on these libraries culminated in the development of predictive models for molecule-level descriptors, as well as the bond- and atom-level descriptors for the conserved reactive site (carboxylic acid or amine). The predictions were confirmed to be robust for an external validation set of medicinally-relevant carboxylic acids and alkyl amines. Additionally, a retrospective study correlating the rate of amide coupling reactions demonstrated the suitability of the predicted DFT-level descriptors for downstream applications. Ultimately, these models enable high-fidelity predictions for a vast number of potential substrates, greatly increasing accessibility to the field of data-driven reaction development. |
| first_indexed | 2025-02-19T04:22:38Z |
| format | Article |
| id | mit-1721.1/158096 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2025-02-19T04:22:38Z |
| publishDate | 2025 |
| publisher | Royal Society of Chemistry |
| record_format | dspace |
| spelling | mit-1721.1/1580962025-01-28T16:55:49Z Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines Haas, Brittany C Hardy, Melissa A Sowndarya S. V., Shree Adams, Keir Coley, Connor W Paton, Robert S Sigman, Matthew S Massachusetts Institute of Technology. Department of Chemical Engineering Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Data-driven reaction discovery and development is a growing field that relies on the use of molecular descriptors to capture key information about substrates, ligands, and targets. Broad adaptation of this strategy is hindered by the associated computational cost of descriptor calculation, especially when considering conformational flexibility. Descriptor libraries can be precomputed agnostic of application to reduce the computational burden of data-driven reaction development. However, as one often applies these models to evaluate novel hypothetical structures, it would be ideal to predict the descriptors of compounds on-the-fly. Herein, we report DFT-level descriptor libraries for conformational ensembles of 8528 carboxylic acids and 8172 alkyl amines towards this goal. Employing 2D and 3D graph neural network architectures trained on these libraries culminated in the development of predictive models for molecule-level descriptors, as well as the bond- and atom-level descriptors for the conserved reactive site (carboxylic acid or amine). The predictions were confirmed to be robust for an external validation set of medicinally-relevant carboxylic acids and alkyl amines. Additionally, a retrospective study correlating the rate of amide coupling reactions demonstrated the suitability of the predicted DFT-level descriptors for downstream applications. Ultimately, these models enable high-fidelity predictions for a vast number of potential substrates, greatly increasing accessibility to the field of data-driven reaction development. 2025-01-28T16:55:47Z 2025-01-28T16:55:47Z 2025-01-15 2025-01-28T16:49:06Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/158096 Haas, Brittany C, Hardy, Melissa A, Sowndarya S. V., Shree, Adams, Keir, Coley, Connor W et al. 2025. "Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines." Digital Discovery, 4 (1). en https://doi.org/10.1039/D4DD00284A Digital Discovery Creative Commons Attribution https://creativecommons.org/licenses/by/4.0/ application/pdf Royal Society of Chemistry Royal Society of Chemistry |
| spellingShingle | Haas, Brittany C Hardy, Melissa A Sowndarya S. V., Shree Adams, Keir Coley, Connor W Paton, Robert S Sigman, Matthew S Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines |
| title | Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines |
| title_full | Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines |
| title_fullStr | Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines |
| title_full_unstemmed | Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines |
| title_short | Rapid prediction of conformationally-dependent DFT-level descriptors using graph neural networks for carboxylic acids and alkyl amines |
| title_sort | rapid prediction of conformationally dependent dft level descriptors using graph neural networks for carboxylic acids and alkyl amines |
| url | https://hdl.handle.net/1721.1/158096 |
| work_keys_str_mv | AT haasbrittanyc rapidpredictionofconformationallydependentdftleveldescriptorsusinggraphneuralnetworksforcarboxylicacidsandalkylamines AT hardymelissaa rapidpredictionofconformationallydependentdftleveldescriptorsusinggraphneuralnetworksforcarboxylicacidsandalkylamines AT sowndaryasvshree rapidpredictionofconformationallydependentdftleveldescriptorsusinggraphneuralnetworksforcarboxylicacidsandalkylamines AT adamskeir rapidpredictionofconformationallydependentdftleveldescriptorsusinggraphneuralnetworksforcarboxylicacidsandalkylamines AT coleyconnorw rapidpredictionofconformationallydependentdftleveldescriptorsusinggraphneuralnetworksforcarboxylicacidsandalkylamines AT patonroberts rapidpredictionofconformationallydependentdftleveldescriptorsusinggraphneuralnetworksforcarboxylicacidsandalkylamines AT sigmanmatthews rapidpredictionofconformationallydependentdftleveldescriptorsusinggraphneuralnetworksforcarboxylicacidsandalkylamines |