Abstract:The introduction of fluorine in compounds plays a crucial role in drug development as it greatly influences their final pharmacokinetic and dynamic properties. Due to the increasing prevalence of fluorine in FDA-approved drugs in recent years, identifying the underlying mechanisms driving their chemical transformations has become crucial in the drug discovery landscape. 19F NMR spectroscopy is a powerful analytical technique that allows for the examination of fluorine-containing compounds, offering valuable information about their structure, dynamics, and reactivity. Consequently, this technique has become a cornerstone in the mechanistic evaluation of fluorine-containing chemical transformations. NMR spectra can be interpreted through the leveraging of Density Functional Theory (DFT), an ab initio modeling method that can be harnessed for the prediction of NMR chemical shifts. However, the screening of compounds and discovery of feasible drug candidates is severely limited due to the computational cost of DFT. Here, we present a machine learning approach to accelerate the prediction of DFT-calculated 19F NMR chemical shifts. The fluorine atoms’ features in the models were derived from their local three-dimensional structural environments, representing their neighboring atoms within a radius of n Å away from the given fluorine atom in the compound. A comparative analysis of thirteen regression models was conducted using features extracted from 501 fluorinated compounds in our laboratory’s chemical inventory. The target chemical shift values were calculated using DFT with the quantum chemistry software ORCA. Among the models evaluated, Gradient Boosting Regression (GBR) exhibited the highest performance, achieving a mean absolute error of 2.89 ppm - 3.73 ppm with a local environment radius of 3 Å. This demonstrates a comparable accuracy to DFT calculations while significantly reducing computational time per compound from several hundred seconds to milliseconds. 3 Å was also found to be the most optimal radius across all models when encoding features for local atomic environments.

Identifying optimal amorphous materials for fluoride removal through Monte Carlo and neural network modeling

Hydrogen Storage Metal-Organic Framework Classification Models Based on Crystal Graph Convolutional Neural Networks

High-Throughput Computation Evaluation of Metal–Organic Frameworks for Efficient Perfluorocarbons Recovery

Aluminum-based Metal Organic Frameworks for Greenhouse Gases CF4 and C2F6 Capture with Excellent Capacity and Selectivity

Prediction of Methane Adsorption Isotherms in Metal–organic Frameworks by Neural Network Synergistic with Classical Density Functional Theory

An Efficient Multi-Scale Lattice Model Approach To Screening Nano-Porous Adsorbents

Integrating Molecular Simulations with Machine Learning Guides in the Design and Synthesis of [BMIM][BF4]/MOF Composites for CO2/N2 Separation

Thermodynamic implications of size, hydrophilicity, and fluorine content on perfluoroalkyl adsorption in NU‐1000

Selective Adsorption of Fluorinated Super Greenhouse Gases within a Metal–Organic Framework with Dynamic Corrugated Ultramicropores

Enhanced adsorption purification of fluoromethane electronic gas using carbon adsorbent with rich ultramicroporosity

Computational screening of metal–organic frameworks for separation of CO2 and N2 from wet flue gas

Process modeling, characterization, optimization, and mechanisms of fluoride adsorption using magnetic agro-based adsorbent

Evaluation of machine learning models for the accelerated prediction of Density Functional Theory calculated 19F chemical shifts based on local atomic environments

Divergent Adsorption Regulation in Metal-Organic Frameworks for Highly Efficient CF4/C2F6 Separation.

Enhanced biosorption of fluoride by extracted nanocellulose/polyvinyl alcohol composite in batch and fixed-bed system: ANN analysis and numerical modeling

Selective Adsorption of the Super Greenhouse Gas Tetrafluoromethane within a Metal–Organic Framework with Dynamic Corrugated Ultramicropores

Machine-learning-based prediction of hydrogen adsorption capacity at varied temperatures and pressures for MOFs adsorbents

Modeling Batch and Column Phosphate Removal by Hydrated Ferric Oxide-Based Nanocomposite Using Response Surface Methodology and Artificial Neural Network

Combining Computational Screening and Machine Learning to Predict Metal–Organic Framework Adsorbents and Membranes for Removing CH4 or H2 from Air

Amino-functionalized novel biosorbent for effective removal of fluoride from water: process optimization using artificial neural network and mechanistic insights

Adsorption Characteristics of Refrigerants for Thermochemical Energy Storage in Metal-Organic Frameworks