Applying statistical modeling strategies to sparse datasets in synthetic chemistry

Science Advances, Год журнала: 2025, Номер 11(1)

Опубликована: Янв. 1, 2025

The application of statistical modeling in organic chemistry is emerging as a standard practice for probing structure-activity relationships and predictive tool many optimization objectives. This review aimed tutorial those entering the area chemistry. We provide case studies to highlight considerations approaches that can be used successfully analyze datasets low data regimes, common situation encountered given experimental demands Statistical hinges on (what being modeled), descriptors (how are represented), algorithms modeled). Herein, we focus how various reaction outputs (e.g., yield, rate, selectivity, solubility, stability, turnover number) structures binned, heavily skewed, distributed) influence choice algorithm constructing chemically insightful models.

Язык: Английский

---

Controlling Stereoselectivity with Noncovalent Interactions in Chiral Phosphoric Acid Organocatalysis

Chemical Reviews, Год журнала: 2025, Номер unknown

Опубликована: Март 18, 2025

Chiral phosphoric acids (CPAs) have emerged as highly effective Brønsted acid catalysts in an expanding range of asymmetric transformations, often through novel multifunctional substrate activation modes. Versatile and broadly appealing, these benefit from modular tunable structures, compatibility with additives. Given the unique types noncovalent interactions (NCIs) that can be established between CPAs various reactants─such hydrogen bonding, aromatic interactions, van der Waals forces─it is unsurprising catalyst systems become a promising approach for accessing diverse chiral product outcomes. This review aims to provide in-depth exploration mechanisms by which impart stereoselectivity, positioning NCIs central feature connects broad spectrum catalytic reactions. Spanning literature 2004 2024, it covers nucleophilic additions, radical atroposelective bond formations, highlighting applicability CPA organocatalysis. Special emphasis placed on structural mechanistic features govern CPA-substrate well tools techniques developed enhance our understanding their behavior. In addition emphasizing details stereocontrolling elements individual reactions, we carefully structured this natural progression specifics broader, class-level perspective. Overall, findings underscore critical role catalysis significant contributions advancing synthesis.

Язык: Английский

---

Materials Genes of CO₂ Hydrogenation on Supported Cobalt Catalysts: An Artificial Intelligence Approach Integrating Theoretical and Experimental Data

Journal of the American Chemical Society, Год журнала: 2024, Номер 146(8), С. 5433 - 5444

Опубликована: Фев. 20, 2024

Designing materials for catalysis is challenging because the performance governed by an intricate interplay of various multiscale phenomena, such as chemical reactions on surfaces and materials' restructuring during catalytic process. In case supported catalysts, role support material can be also crucial. Here, we address this intricacy challenge a symbolic-regression artificial intelligence (AI) approach. We identify key physicochemical parameters correlated with measured performance, out many offered candidate characterizing materials, reaction environment, possibly relevant underlying phenomena. Importantly, these are obtained both experiments ab initio simulations. The identified might called "materials genes", in analogy to genes biology: they correlate property or function interest, but explicit physical relationship not (necessarily) known. To demonstrate approach, investigate CO2 hydrogenation catalyzed cobalt nanoparticles silica. Crucially, silica modified additive metals magnesium, calcium, titanium, aluminum, zirconium, which results six significantly different performances. These systems mimic hydrothermal vents, have produced first organic molecules Earth. CH3OH selectivity reflect reducibility species, adsorption strength intermediates, nature metal. By using AI model trained basic elemental properties (e.g., ionization potential) parameters, new additives suggested. predicted catalysts vanadium zinc confirmed experiments.

Язык: Английский

---

Designing Target-specific Data Sets for Regioselectivity Predictions on Complex Substrates

Journal of the American Chemical Society, Год журнала: 2025, Номер 147(9), С. 7476 - 7484

Опубликована: Фев. 21, 2025

The development of machine learning models to predict the regioselectivity C(sp3)-H functionalization reactions is reported. A data set for dioxirane oxidations was curated from literature and used generate a model C-H oxidation. To assess whether smaller, intentionally designed sets could provide accuracy on complex targets, series acquisition functions were developed select most informative molecules specific target. Active learning-based that leverage predicted reactivity uncertainty found outperform those based molecular site similarity alone. use elaboration significantly reduced number points needed perform accurate prediction, it machine-designed can give predictions when larger, randomly selected fail. Finally, workflow experimentally validated five substrates shown be applicable predicting arene radical borylation. These studies quantitative alternative intuitive extrapolation "model substrates" frequently estimate molecules.

Язык: Английский

---

Chemoinformatic Catalyst Selection Methods for the Optimization of Copper–Bis(oxazoline)-Mediated, Asymmetric, Vinylogous Mukaiyama Aldol Reactions

ACS Catalysis, Год журнала: 2024, Номер 14(4), С. 2642 - 2655

Опубликована: Фев. 6, 2024

A catalyst selection method for the optimization of an asymmetric, vinylogous Mukaiyama aldol reaction is described. large library commercially available and synthetically accessible copper–bis(oxazoline) catalysts was constructed in silico. Conformer-dependent, grid-based descriptors were calculated each catalyst, defining a chemical feature space suitable machine learning. Selection diverse subset produced initial training set 26 new bis(oxazoline) ligands that synthesized tested stereoselectivity copper-catalyzed, five substrate combinations. One ligand provided 88% average enantiomeric excess, exceeding performance identified through campaign. Supervised unsupervised methods, including quantitative structure–selectivity relationship modeling, nearest neighbors analysis, focused analogue clustering strategy, employed to identify additional 12 ligands. The selected outperformed hit four out product classes some cases demonstrated enantiocontrol 95% ee. effectiveness process discussed, expediency neighbor approaches are contrasted with supervised modeling approach.

Язык: Английский

---

C–H Aminoalkylation of 5-Membered Heterocycles: Influence of Descriptors, Data Set Size, and Data Quality on the Predictiveness of Machine Learning Models and Expansion of the Substrate Space Beyond 1,3-Azoles

The Journal of Organic Chemistry, Год журнала: 2025, Номер unknown

Опубликована: Фев. 11, 2025

We report a general C-H aminoalkylation of 5-membered heterocycles through combined machine learning/experimental workflow. Our work describes previously unknown functionalization reactivity and creates predictive learning (ML) model iterative refinement over 6 rounds active learning. The initial established with 1,3-azoles predicts the reactivities N-aryl indazoles, 1,2,4-triazolopyrazines, 1,2,3-thiadiazoles, 1,3,4-oxadiazoles, while other substrate classes (e.g., pyrazoles 1,2,4-triazoles) are not predicted well. final includes additional heterocyclic scaffolds in training data, which results high accuracy across all tested cores. prediction performance is shown both within set via cross-validation (CV R2 = 0.81) when predicting unseen substrates diverse molecular weight structure (Test 0.95). concept feature engineering discussed, we benchmark mechanistically related DFT-based features that more time-intensive laborious comparison descriptors fingerprints. Importantly, this establishes novel for methods underdeveloped. Since such key motifs drug discovery development, expect to be significant use synthetic synthesis-oriented ML communities.

Язык: Английский

---

Probability Guided Chemical Reaction Scopes

Опубликована: Янв. 1, 2025

Язык: Английский

---

Computational Tools for the Prediction of Site- and Regioselectivity of Organic Reactions

Chemical Science, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

This article reviews computational tools for the prediction of regio- and site-selectivity organic reactions. It spans from quantum chemical procedures to deep learning models showcases application presented tools.

Язык: Английский

---

Optimising Materials Properties with Minimal Data: Lessons from Vanadium Catalyst Modelling

Challenges and advances in computational chemistry and physics, Год журнала: 2025, Номер unknown, С. 117 - 138

Опубликована: Янв. 1, 2025

Язык: Английский

---

Machine learning workflows beyond linear models in low-data regimes

Chemical Science, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

This work presents automated non-linear workflows for studying problems in low-data regimes alongside traditional linear models.

Язык: Английский

---