An Unsupervised Machine Learning Workflow for Assigning and Predicting Generality in Asymmetric Catalysis DOI Creative Commons
Isaiah O. Betinol,

Saumya Thakur,

Jolene P. Reid

et al.

Published: Dec. 16, 2022

The development of chiral catalysts that can provide high enantioselectivities across a wide assortment substrates or reaction range is priority for many catalyst design efforts. While several approaches are available to aid in the identification general systems there currently no simple procedure directly measuring how given could be. Herein, we present catalyst-agnostic workflow centered on unsupervised machine learning enables rapid assessment and quantification generality. uses curated literature data sets descriptors visualize cluster chemical space coverage. This network then be applied derive generality metric through designer equations interfaced with other regression techniques prediction. As validating case studies, have successfully this method identify-through-quantification most chemotype an organocatalytic asymmetric Mannich predicted phosphoric acid addition nucleophile imines. mechanistic basis gleaned from calculated values by deconstructing contributions enantiomeric excess overall result. We conclude broadly applicable may more adaptative changes reactant structure because enantioinduction does not rely single set noncovalent interactions. In contrast, some work engaging robust contacts do change significantly nature when component altered. Ultimately, our findings represent framework interrogating predicting generality, strategy should relevant catalytic widely synthesis.

Language: Английский

Towards a comprehensive data infrastructure for redox-active organic molecules targeting non-aqueous redox flow batteries DOI Creative Commons
Rebekah Duke, Vinayak Bhat, Parker Sornberger

et al.

Digital Discovery, Journal Year: 2023, Volume and Issue: 2(4), P. 1152 - 1162

Published: Jan. 1, 2023

The D 3 TaLES database and data infrastructure aim to offer readily accessible uniform of varying types for redox-active organic molecules targeting non-aqueous redox flow batteries.

Language: Английский

Citations

9
Automated prediction of ground state spin for transition metal complexes DOI Creative Commons
Yuri Cho, Rubén Laplaza, Sergi Vela

et al.

Digital Discovery, Journal Year: 2024, Volume and Issue: 3(8), P. 1638 - 1647

Published: Jan. 1, 2024

Exploiting crystallographic data repositories for large-scale quantum chemical computations requires the rapid and accurate extraction of molecular structure, charge spin from information file. Here, we develop a general approach to assign ground state transition metal complexes, in complement our previous efforts on determining oxidation states bond order within

Language: Английский

Citations

3
Recent advances of machine learning applications in the development of experimental homogeneous catalysis DOI Creative Commons

Nil Sanosa,

David Dalmau, Diego Sampedro

et al.

Artificial Intelligence Chemistry, Journal Year: 2024, Volume and Issue: 2(1), P. 100068 - 100068

Published: April 27, 2024

Language: Английский

Citations

2
Catalysing (organo-)catalysis: Trends in the application of machine learning to enantioselective organocatalysis DOI Creative Commons
Stefan P. Schmid, Leon Schlosser, Frank Glorius

et al.

Beilstein Journal of Organic Chemistry, Journal Year: 2024, Volume and Issue: 20, P. 2280 - 2304

Published: Sept. 10, 2024

Organocatalysis has established itself as a third pillar of homogeneous catalysis, besides transition metal catalysis and biocatalysis, its use for enantioselective reactions gathered significant interest over the last decades. Concurrent to this development, machine learning (ML) been increasingly applied in chemical domain efficiently uncover hidden patterns data accelerate scientific discovery. While uptake ML organocatalysis comparably slow, two decades have showed an increased from community. This review gives overview work field organocatalysis. The starts by giving short primer on experimental chemists, before discussing application predicting selectivity organocatalytic transformations. Subsequently, we employed privileged catalysts, focusing catalyst reaction design. Concluding, give our view current challenges future directions field, drawing inspiration other domains.

Language: Английский

Citations

2
Computational methods for asymmetric catalysis DOI
Sharon Pinus, Jérôme Genzling, Mihai Burai Patrascu

et al.

Nature Catalysis, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 3, 2024

Language: Английский

Citations

2
Classification of Hemilabile Ligands Using Machine Learning DOI
Ilia Kevlishvili, Chenru Duan, Heather J. Kulik

et al.

The Journal of Physical Chemistry Letters, Journal Year: 2023, Volume and Issue: 14(49), P. 11100 - 11109

Published: Dec. 5, 2023

Hemilabile ligands have the capacity to partially disengage from a metal center, providing strategy balance stability and reactivity in catalysis, but they are not straightforward identify. We identify Cambridge Structural Database that been crystallized with distinct denticities thus identifiable as hemilabile ligands. implement semi-supervised learning approach using label-spreading algorithm augment small negative set is supported by heuristic rules of ligand co-occurrence. show based on coordinating atom identity alone sufficient whether hemilabile, our trained machine-learning classification models instead needed predict bi-, tri-, or tetradentate high accuracy precision. Feature importance analysis shows second, third, fourth coordination spheres all play important roles hemilability.

Language: Английский

Citations

6
Combining Molecular Quantum Mechanical Modeling and Machine Learning for Accelerated Reaction Screening and Discovery DOI
Nicholas Casetti, Javier Emilio Alfonso Ramos,

Connor W. Coley

et al.

Chemistry - A European Journal, Journal Year: 2023, Volume and Issue: 29(60)

Published: Aug. 1, 2023

Molecular quantum mechanical modeling, accelerated by machine learning, has opened the door to high-throughput screening campaigns of complex properties, such as activation energies chemical reactions and absorption/emission spectra materials molecules; in silico. Here, we present an overview main principles, concepts, design considerations involved hybrid computational chemistry/machine learning workflows, with a special emphasis on some recent examples their successful application. We end brief outlook further advances that will benefit field.

Language: Английский

Citations

5
Identification via Virtual Screening of Emissive Molecules with a Small Exciton–Vibration Coupling for High Color Purity and Potential Large Exciton Delocalization DOI Creative Commons
Xiaoyu Xie, Alessandro Troisi

The Journal of Physical Chemistry Letters, Journal Year: 2023, Volume and Issue: 14(17), P. 4119 - 4126

Published: April 27, 2023

A sequence of quantum chemical computations increasing accuracy was used in this work to identify molecules with small exciton reorganization energy (exciton–vibration coupling), interest for light emitting devices and coherent transport, starting from a set ∼4500 known molecules. We validated an approximate computational approach based on single-point calculations the force excited state, which shown be very efficient identifying most promising candidates. showed that simple descriptor bond order could find potentially energies without performing state calculations. chemically diverse analyzed greater detail common features leading property. Many such display A–B–A structure where bonding/antibonding patterns fragments are similar HOMO LUMO. Another group displays instead LUMO strong nonbonding character.

Language: Английский

Citations

4
From Organic Fragments to Photoswitchable Catalysts: The OFF–ON Structural Repository for Transferable Kernel-Based Potentials DOI Creative Commons
Frédéric Célerse, Matthew D. Wodrich, Sergi Vela

et al.

Journal of Chemical Information and Modeling, Journal Year: 2024, Volume and Issue: 64(4), P. 1201 - 1212

Published: Feb. 6, 2024

Structurally and conformationally diverse databases are needed to train accurate neural networks or kernel-based potentials capable of exploring the complex free energy landscape flexible functional organic molecules. Curating such for species beyond "simple" drug-like compounds molecules composed well-defined building blocks (e.g., peptides) is challenging as it requires thorough chemical space mapping evaluation both conformational diversities. Here, we introduce OFF-ON (organic fragments from organocatalysts that non-modular) database, a repository 7869 equilibrium 67,457 nonequilibrium geometries dimers aimed at describing molecules, with an emphasis on photoswitchable organocatalysts. The relevance this database then demonstrated by training local kernel regression model low-cost semiempirical baseline comparing PBE0-D3 reference several known catalysts, notably surfaces exemplary Our results demonstrate data set offers reliable predictions simulating behavior virtually any (photoswitchable) organocatalyst compound H, C, N, O, F, S atoms, thereby opening computationally feasible route explore in order rationalize predict catalytic behavior.

Language: Английский

Citations

1
Precise Clearance of Intracellular MRSA via Internally and Externally Mediated Bioorthogonal Activation of Micro/Nano Hydrogel Microspheres DOI Creative Commons
Jianye Yang, Li Chen, Zhengwei Cai

et al.

Advanced Science, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 29, 2024

Abstract Traditional high‐dose antibiotic treatments of intracellular methicillin‐resistant staphylococcus aureus (MRSA) are highly inefficient and associated with a high rate infection relapse. As an effective antibacterial technology, sonodynamic therapy (SDT) may be able to break the dilemma. However, indiscriminate reactive oxygen species (ROS) release leads potential side effects. This study incorporates Staphylococcal Protein A antibody‐modified Cu 2+ /tetracarboxyphenylporphyrin nanoparticles (Cu(II)NS‐SPA) into hydrogel microspheres (HAMA@Cu(II)NS‐SPA) achieve precise eradication bacteria. is under bioorthogonal activation mediated by bacillithiol (BSH) (internally) ultrasound (US) (externally). To specify, US responsiveness Cu(II)NS‐SPA restored when it reduced Cu(I)NS‐SPA BSH secreted characteristically MRSA, thus forming external US, which confines ROS production within infected M Φ . Under at 2 W cm −2 , over 95% MRSA can cleared. In vivo, single injection HAMA@Cu(II)NS‐SPA achieves up two weeks therapy, reducing pro‐inflammatory factor expression 90%, peri‐implant bone trabeculae numbers exceed control group five times. summary, these micro/nano internal precisely eliminate effectively treating multi‐drug resistant bacterial infections.

Language: Английский

Citations

1