A5093979178- Machine Learning in Materials Science
- Topic Modeling
- Chemical Synthesis and Analysis
- Asymmetric Hydrogenation and Catalysis
- Asymmetric Synthesis and Catalysis
- Computational Drug Discovery Methods
- Intermetallics and Advanced Alloy Properties
- Advanced Graph Neural Networks
- Nuclear Materials and Properties
- Metal and Thin Film Mechanics
University of Nottingham
2024-2025
The range of chemical databases available has dramatically increased in recent years, but the reliability and quality their data are often negatively affected by human-error fidelity. size can make manual curation/checking such sets time consuming; thus, automated tools to help this process highly desirable. Herein, we propose use Graph Neural Networks (GNNs) identifying potential stereochemical misassignments primary asymmetric catalysis literature. Our method relies on an ensemble GNN...
Homogeneous catalysts enable faster conversions of molecules with higher selectivities (stereo- and regioselectivity) in chemical reactions. Traditionally, catalyst improvements are made through empirical trials, where the is functionalised by adding, removing or modifying groups within its structure and, subsequently, reevaluating new catalytic activity. This procedure not efficient leads to unsuccessful trials that waste resources. Machine learning (ML) approaches have been proposed...
Despite its current popularity, machine learning (ML) applied to asymmetric catalysis remains underexplored. Present strategies include direct use of existing descriptors (e.g. those originally formulated for medicinal chemistry), the development new bespoke steric and electronic descriptors, or molecular graphs. This method diversity, in absence user guidelines, makes selecting an optimal ML algorithm unclear. The fact that data sets are frequently small also make interpretable chiral...
Despite its current popularity, machine learning (ML) applied to asymmetric catalysis remains underexplored. Present strategies include direct use of existing descriptors (e.g. those originally formulated for medicinal chemistry), the development new bespoke steric and electronic descriptors, or molecular graphs. This method diversity, in absence user guidelines, makes selecting an optimal ML algorithm unclear. The fact that data sets are frequently small also make interpretable chiral...
<title>Abstract</title> The properties of periodic solid-state materials are frequently modified by the inclusion interstitial atoms deposited pseudo-randomly throughout crystal lattice. Accurately calculating overall lattice currently requires time-intensive, high-level quantum (typically DFT) calculations, often on 1000s stochastic lattices. 'Interatomic potential models' (IPM) can mitigate such computational burdens awarding discrete coefficients to common geometric ensembles within wider...