Analyzing the Accuracy of Critical Micelle Concentration Predictions Using Deep Learning DOI
Alexander Moriarty, Takeshi Kobayashi, Matteo Salvalaglio

et al.

Journal of Chemical Theory and Computation, Journal Year: 2023, Volume and Issue: 19(20), P. 7371 - 7386

Published: Oct. 10, 2023

This paper presents a novel approach to predicting critical micelle concentrations (CMCs) by using graph neural networks (GNNs) augmented with Gaussian processes (GPs). The proposed model uses learned latent space representations of molecules predict CMCs and estimate uncertainties. performance the on data set containing nonionic, cationic, anionic, zwitterionic is compared against linear that works extended connectivity fingerprints (ECFPs). GNN-based performs slightly better than ECFP when there enough well-balanced training achieves predictive accuracy comparable published models were evaluated smaller range surfactant chemistries. We illustrate applicability domain our molecular cartogram visualize space, which helps identify for predictions are likely be erroneous. In addition accurately some classes, can provide valuable insights into properties influence CMCs.

Language: Английский

Analyzing the Accuracy of Critical Micelle Concentration Predictions Using Deep Learning DOI
Alexander Moriarty, Takeshi Kobayashi, Matteo Salvalaglio

et al.

Journal of Chemical Theory and Computation, Journal Year: 2023, Volume and Issue: 19(20), P. 7371 - 7386

Published: Oct. 10, 2023

This paper presents a novel approach to predicting critical micelle concentrations (CMCs) by using graph neural networks (GNNs) augmented with Gaussian processes (GPs). The proposed model uses learned latent space representations of molecules predict CMCs and estimate uncertainties. performance the on data set containing nonionic, cationic, anionic, zwitterionic is compared against linear that works extended connectivity fingerprints (ECFPs). GNN-based performs slightly better than ECFP when there enough well-balanced training achieves predictive accuracy comparable published models were evaluated smaller range surfactant chemistries. We illustrate applicability domain our molecular cartogram visualize space, which helps identify for predictions are likely be erroneous. In addition accurately some classes, can provide valuable insights into properties influence CMCs.

Language: Английский

Citations

10