Transferability Across Different Molecular Systems and Levels of Theory with the Data-Driven Coupled-Cluster Scheme DOI

P. D. Varuna S. Pathirage,

Brody Quebedeaux,

Shahzad Akram

et al.

The Journal of Physical Chemistry A, Journal Year: 2025, Volume and Issue: unknown

Published: March 25, 2025

Machine learning has recently been introduced into the arsenal of tools that are available to computational chemists. In past few years, we have seen an increase in applicability these on a plethora applications, including automated exploration large fraction chemical space, reduction repetitive tasks, detection outliers databases, and acceleration molecular simulations. An attractive application machine electronic structure theory is "recycling" wave functions for faster more accurate completion complex quantum calculations. Along lines, developed hybrid chemical/machine workflows utilize information from low-level prediction higher-level functions. The data-driven coupled-cluster (DDCC) family methods discussed this article together with importance inclusion physical properties such workflows. After short introduction philosophy capabilities DDCC, present our recent progress extending its larger structures data sets. A significant advantage offered by DDCC transferability, respect different systems excitation levels. As show here, predicted at singles doubles level can be used perturbative triples CCSD(T) scheme. We conclude some personal considerations future directions related development next generation models.

Language: Английский

Designing semiconductor materials and devices in the post-Moore era by tackling computational challenges with data-driven strategies DOI
Jiahao Xie, Yansong Zhou, Muhammad Faizan

et al.

Nature Computational Science, Journal Year: 2024, Volume and Issue: 4(5), P. 322 - 333

Published: May 23, 2024

Language: Английский

Citations

7
Ligand-Based Principal Component Analysis Followed by Ridge Regression: Application to an Asymmetric Negishi Reaction DOI
H. Ray Kelly, Sanil Sreekumar, Vidhyadhar Manee

et al.

ACS Catalysis, Journal Year: 2024, Volume and Issue: 14(7), P. 5027 - 5038

Published: March 20, 2024

In this study, we introduce an approach for predicting the enantioselectivity of P-chiral monophosphorus ligands from ligand-based descriptors that can be applied to catalytic systems with small experimental datasets without reliance on mechanistic knowledge. Principal component analysis (PCA) is used map out chemical space described by steric and electronic computed dihydrobenzooxaphosphole (BOP) dihydrobenzoazaphosphole (BAP) ligands. The PCA captures trends in experimentally measured four C–C bond-forming reactions identifies "hotspots" selective provide insight into optimal balance sterics electronics each reaction. Furthermore, are train a ridge regression model quantitatively predicts Pd-catalyzed Negishi cross-coupling coefficients fundamental understanding reveal π-stacking interaction one results unexpected selectivity inversion. Overall, integrated combines qualitative quantitative (ridge regression) predictions.

Language: Английский

Citations

6
ROBERT: Bridging the Gap Between Machine Learning and Chemistry DOI Creative Commons
David Dalmau, Juan V. Alegre‐Requena

Wiley Interdisciplinary Reviews Computational Molecular Science, Journal Year: 2024, Volume and Issue: 14(5)

Published: Sept. 1, 2024

ABSTRACT Beyond addressing technological demands, the integration of machine learning (ML) into human societies has also promoted sustainability through adoption digitalized protocols. Despite these advantages and abundance available toolkits, a substantial implementation gap is preventing widespread incorporation ML protocols computational experimental chemistry communities. In this work, we introduce ROBERT, software carefully crafted to make more accessible chemists all programming skill levels, while achieving results comparable those field experts. We conducted benchmarking using six recent studies in containing from 18 4149 entries. Furthermore, demonstrated program's ability initiate workflows directly SMILES strings, which simplifies generation predictors for common problems. To assess ROBERT's practicality real‐life scenarios, employed it discover new luminescent Pd complexes with modest dataset 23 points, frequently encountered scenario studies.

Language: Английский

Citations

6
OSCAR: an extensive repository of chemically and functionally diverse organocatalysts DOI Creative Commons
Simone Gallarati, Puck van Gerwen, Rubén Laplaza

et al.

Chemical Science, Journal Year: 2022, Volume and Issue: 13(46), P. 13782 - 13794

Published: Jan. 1, 2022

A database of thousands experimentally-derived or combinatorially enriched organocatalysts and fragments to navigate chemical space optimize reaction properties.

Language: Английский

Citations

25
It Is Not All about the Ligands: Exploring the Hidden Potentials of tBu3P through Its Oxidative Addition Complex as the Precatalyst DOI

Yam N. Timsina,

Guolin Xu, Thomas J. Colacot

et al.

ACS Catalysis, Journal Year: 2023, Volume and Issue: 13(12), P. 8106 - 8118

Published: June 1, 2023

A series of oxidative addition complexes with a general formula (tBu3P)Pd(Ar)X, as class precatalysts, were synthesized for challenging Suzuki–Miyaura coupling involving partners, such (i) sensitive polyfluorinated arylboronic acids or their corresponding boronic esters, (ii) sterically hindered electrophiles, and (iii) nucleophiles. total 89 examples are reported, which 39 in the Supporting Information. These particular (tBu3P)Pd(4-CF3Ph)Br, demonstrated to be powerful catalytic systems cross reactions comparison situ created by mixing tBu3P ligand palladium precursor. The precatalysts also superior other monoligated systems, Buchwald's biaryl based G3 G4 palladacycles. In addition, (tBu3P)Pd(4-CF3Ph)Br precatalyst is highly effective second most popular reaction, namely Buchwald–Hartwig coupling. this study, electron-deficient amines coupled (hetero)aryl bromides chlorides 34 examples, 8 Interestingly, results obtained both C–C C–N couplings on par that "state-of-the-art" catalysts containing Ad3P Np3P ligands same similar substrates, suggesting it not all about ligands.

Language: Английский

Citations

14
Chemoinformatic Catalyst Selection Methods for the Optimization of Copper–Bis(oxazoline)-Mediated, Asymmetric, Vinylogous Mukaiyama Aldol Reactions DOI
Casey L. Olen, Andrew F. Zahrt, Sean W. Reilly

et al.

ACS Catalysis, Journal Year: 2024, Volume and Issue: 14(4), P. 2642 - 2655

Published: Feb. 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.

Language: Английский

Citations

5
Automated approaches, reaction parameterisation, and data science in organometallic chemistry and catalysis: towards improving synthetic chemistry and accelerating mechanistic understanding DOI Creative Commons
Stuart C. Smith, Christopher S. Horbaczewskyj, Theo F. N. Tanner

et al.

Digital Discovery, Journal Year: 2024, Volume and Issue: 3(8), P. 1467 - 1495

Published: Jan. 1, 2024

This review discusses the use of automation for organometallic reactions to generate rich datasets and, with statistical analysis and reaction component parameterisation, how mechanisms can be probed gain understanding.

Language: Английский

Citations

4
Data-science-guided calibration curve prediction of an MLCT-based ee determination assay for chiral amines DOI
James R. Howard, Julia R. Shuluk,

Arya Bhakare

et al.

Chem, Journal Year: 2024, Volume and Issue: 10(7), P. 2074 - 2088

Published: June 12, 2024

Language: Английский

Citations

4
Multi‐Threshold Analysis for Chemical Space Mapping of Ni‐Catalyzed Suzuki‐Miyaura Couplings DOI

Austin LeSueur,

Nari Tao,

Abigail G. Doyle

et al.

European Journal of Organic Chemistry, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 11, 2024

Abstract A key challenge in synthetic chemistry is the selection of high‐performing ligands for cross‐coupling reactions. To address this challenge, work presents a classification workflow to identify physicochemical descriptors that bin monophosphine as active or inactive Ni‐catalyzed Suzuki‐Miyaura coupling Using five previously published high‐throughput experimentation datasets training, we found binary classifier using phosphine's minimum buried volume and Boltzmann‐averaged electrostatic potential most effective at distinguishing high low‐yielding ligands. Experimental validations are also presented. two from represent chemical space leads more predictive guide structure‐reactivity relationships compared with classic representations.

Language: Английский

Citations

4
COBRA web application to benchmark linear regression models for catalyst optimization with few-entry datasets DOI Creative Commons
Zhen Cao, Laura Falivene, Albert Poater

et al.

Cell Reports Physical Science, Journal Year: 2024, Volume and Issue: unknown, P. 102348 - 102348

Published: Dec. 1, 2024

Language: Английский

Citations

4