| Autor: |
Borup RM; Department of Chemistry, University of Copenhagen, Copenhagen, DK-2100, Denmark., Ree N; Department of Chemistry, University of Copenhagen, Copenhagen, DK-2100, Denmark., Jensen JH; Department of Chemistry, University of Copenhagen, Copenhagen, DK-2100, Denmark. |
| Jazyk: |
angličtina |
| Zdroj: |
Beilstein journal of organic chemistry [Beilstein J Org Chem] 2024 Jul 16; Vol. 20, pp. 1614-1622. Date of Electronic Publication: 2024 Jul 16 (Print Publication: 2024). |
| DOI: |
10.3762/bjoc.20.144 |
| Abstrakt: |
Determining the p K a values of various C-H sites in organic molecules offers valuable insights for synthetic chemists in predicting reaction sites. As molecular complexity increases, this task becomes more challenging. This paper introduces pKalculator, a quantum chemistry (QM)-based workflow for automatic computations of C-H p K a values, which is used to generate a training dataset for a machine learning (ML) model. The QM workflow is benchmarked against 695 experimentally determined C-H p K a values in DMSO. The ML model is trained on a diverse dataset of 775 molecules with 3910 C-H sites. Our ML model predicts C-H p K a values with a mean absolute error (MAE) and a root mean squared error (RMSE) of 1.24 and 2.15 p K a units, respectively. Furthermore, we employ our model on 1043 p K a -dependent reactions (aldol, Claisen, and Michael) and successfully indicate the reaction sites with a Matthew's correlation coefficient (MCC) of 0.82.
Competing Interests: The authors declare that there are no competing interests.
(Copyright © 2024, Borup et al.) |
| Databáze: |
MEDLINE |
| Externí odkaz: |
|
