The computational road to reactivity scales

Vahl, Maike; Proppe, Jonny

doi:10.1039/d2cp03937k

Cited by 12 publications

(18 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It can be seen from Figure 4c that in most cases, the predicted error of r SPOC model was within±2.0 in different solvents including those mixed solvents. What's more, in Mayr's database, the nonuniformity of errors was qualitatively considered through a star‐rating system [3e] . Proppe et al.…”

Section: Resultsmentioning

confidence: 99%

Prediction of Nucleophilicity and Electrophilicity Based on a Machine‐Learning Approach

Liu

Cheng

et al. 2023

ChemPhysChem

View full text Add to dashboard Cite

Nucleophilicity and electrophilicity dictate the reactivity of polar organic reactions. In the past decades, Mayr et al. established a quantitative scale for nucleophilicity (N) and electrophilicity (E), which proved to be a useful tool for the rationalization of chemical reactivity. In this study, a holistic prediction model was developed through a machine-learning approach. rSPOC, an ensemble molecular representation with structural, physicochemical and solvent features, was developed for this purpose. With 1115 nucleophiles, 285 electrophiles, and 22 solvents, the dataset is currently the largest one for reactivity prediction. The rSPOC model trained with the Extra Trees algorithm showed high accuracy in predicting Mayr's N and E parameters with R 2 of 0.92 and 0.93, MAE of 1.45 and 1.45, respectively. Furthermore, the practical applications of the model, for instance, nucleophilicity prediction of NADH, NADPH and a series of enamines showed potential in predicting molecules with unknown reactivity within seconds. An online prediction platform (http://isyn.luoszgroup.com/) was constructed based on the current model, which is available free to the scientific community.

show abstract

Section: Resultsmentioning

confidence: 99%

Prediction of Nucleophilicity and Electrophilicity Based on a Machine‐Learning Approach

Liu

Cheng

et al. 2023

ChemPhysChem

View full text Add to dashboard Cite

show abstract

“…This is one reason why data-driven or machine-learning (ML) algorithms have gained much attention in this context as they are capable of yielding fast predictions by interpolating between available data. 14 In supervised ML, relationships between descriptors (input variables) and targets (output variables) are learned by means of regression (continuous target) or classification (discrete target). Aside from the expensive acquisition of targets (i.e., experimental reactivity parameters), the generation of descriptors can constitute a critical bottleneck in the ML workflow.…”

Section: ■ Introductionmentioning

confidence: 99%

“…As they proceeded, Mayr and his team found that the Mayr–Patz equation (eq ) is also valid for many other classes of nucleophiles and electrophiles. To date, reactivity parameters have been determined for 352 electrophiles ( E ) and 1281 nucleophiles ( N , s N ), which can be accessed via Mayr’s Database of Reactivity Parameters . , A brief explanation of how these parameters are determined experimentally , is given in boxes 1 and 2 of ref .…”

Section: Introductionmentioning

confidence: 99%

“…Our group currently explores the suitability of quantitative structure–reactivity relationships (QSRRs) for accelerating reactivity predictions and, hence, synthesis planning. − We aspire to build an interactive platform on which users can query arbitrary organic compounds and receive instant feedback, including site-specific reactivity information and uncertainty estimates to ensure reliability and practical benefit. With the ability to assess reactivity in real time, chemists can efficiently evaluate a vast number of compounds and potential reactions and choose the most promising ones for further investigation.…”

Section: Introductionmentioning

confidence: 99%

“…Ultimately, DFT is not suitable for the efficient prediction of the reactivity parameters. This is one reason why data-driven or machine-learning (ML) algorithms have gained much attention in this context as they are capable of yielding fast predictions by interpolating between available data …”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Quantitative Structure–Reactivity Relationships for Synthesis Planning: The Benzhydrylium Case

Eckhoff,

Diedrich,

Mücke

et al. 2023

J. Phys. Chem. A

Self Cite

View full text Add to dashboard Cite

Selective and feasible reactions are among the top targets in synthesis planning. Mayr's approach to quantifying chemical reactivity has greatly facilitated the planning process, but reactivity parameters for new compounds require time-consuming experiments. In the past decade, data-driven modeling has been gaining momentum in the field, as it shows promise in terms of efficient reactivity prediction. However, state-of-the-art models use quantum chemical data as input, which prevent access to real-time planning in organic synthesis. Here, we present a novel data-driven workflow for predicting reactivity parameters of molecules that takes only structural information as input, enabling de facto realtime reactivity predictions. We use the well-understood chemical space of benzhydrylium ions as an example to demonstrate the functionality of our approach and the performance of the resulting quantitative structure−reactivity relationships (QSRRs). Our results suggest that it is straightforward to build low-cost QSRR models that are accurate, interpretable, and transferable to unexplored systems within a given scope of application. Moreover, our QSRR approach suggests that Hammett σ parameters are only approximately additive.

show abstract

Revisiting the Paradigm of Reaction Optimization in Flow with a Priori Computational Reaction Intelligence

Bianchi,

Monbaliu

2023

Angewandte Chemie

View full text Add to dashboard Cite

The use of micro/meso‐fluidic reactors has resulted in both new scenarios for chemistry and new requirements for chemists. Through flow chemistry, large‐scale reactions can be performed in drastically reduced reactor sizes and reaction times. This obvious advantage comes with the concomitant challenge of re‐designing long‐established batch processes to fit these new conditions. The reliance on experimental trial‐and‐error to perform this translation frequently makes flow chemistry unaffordable, thwarting initial aspirations to revolutionize chemistry. By combining computational chemistry and machine learning, we have developed a model that provides predictive power tailored specifically to flow reactions. We show its applications to translate batch to flow, provide mechanistic insight, contribute reagent descriptors, and to synthesize a library of novel compounds in excellent yields after executing a single set of conditions.

show abstract

The computational road to reactivity scales

Abstract: Reactivity scales are powerful research tools. This tutorial shows how to create and use them on the computer.

Cited by 12 publications

References 71 publications

Prediction of Nucleophilicity and Electrophilicity Based on a Machine‐Learning Approach

Prediction of Nucleophilicity and Electrophilicity Based on a Machine‐Learning Approach

Quantitative Structure–Reactivity Relationships for Synthesis Planning: The Benzhydrylium Case

Revisiting the Paradigm of Reaction Optimization in Flow with a Priori Computational Reaction Intelligence

Contact Info

Product

Resources

About