Prediction of higher-selectivity catalysts by computer-driven workflow and machine learning

Andrew F. Zahrt, Jeremy J. Henle, Brennan T. Rose, Yang Wang, William T. Darrow, Scott E. Denmark

Research output: Contribution to journalArticle

Abstract

Catalyst design in asymmetric reaction development has traditionally been driven by empiricism, wherein experimentalists attempt to qualitatively recognize structural patterns to improve selectivity. Machine learning algorithms and chemoinformatics can potentially accelerate this process by recognizing otherwise inscrutable patterns in large datasets. Herein we report a computationally guided workflow for chiral catalyst selection using chemoinformatics at every stage of development. Robust molecular descriptors that are agnostic to the catalyst scaffold allow for selection of a universal training set on the basis of steric and electronic properties. This set can be used to train machine learning methods to make highly accurate predictive models over a broad range of selectivity space. Using support vector machines and deep feed-forward neural networks, we demonstrate accurate predictive modeling in the chiral phosphoric acid–catalyzed thiol addition to N-acylimines.

Original languageEnglish (US)
Article numbereaau5631
JournalScience
Volume363
Issue number6424
DOIs
StatePublished - Jan 18 2019

Fingerprint

Workflow
Empiricism
Sulfhydryl Compounds
Machine Learning
Support Vector Machine
Datasets

ASJC Scopus subject areas

  • General

Cite this

Prediction of higher-selectivity catalysts by computer-driven workflow and machine learning. / Zahrt, Andrew F.; Henle, Jeremy J.; Rose, Brennan T.; Wang, Yang; Darrow, William T.; Denmark, Scott E.

In: Science, Vol. 363, No. 6424, eaau5631, 18.01.2019.

Research output: Contribution to journalArticle

Zahrt, Andrew F. ; Henle, Jeremy J. ; Rose, Brennan T. ; Wang, Yang ; Darrow, William T. ; Denmark, Scott E. / Prediction of higher-selectivity catalysts by computer-driven workflow and machine learning. In: Science. 2019 ; Vol. 363, No. 6424.
@article{3f560f592e7d4ad692fe44b9dcefc4a5,
title = "Prediction of higher-selectivity catalysts by computer-driven workflow and machine learning",
abstract = "Catalyst design in asymmetric reaction development has traditionally been driven by empiricism, wherein experimentalists attempt to qualitatively recognize structural patterns to improve selectivity. Machine learning algorithms and chemoinformatics can potentially accelerate this process by recognizing otherwise inscrutable patterns in large datasets. Herein we report a computationally guided workflow for chiral catalyst selection using chemoinformatics at every stage of development. Robust molecular descriptors that are agnostic to the catalyst scaffold allow for selection of a universal training set on the basis of steric and electronic properties. This set can be used to train machine learning methods to make highly accurate predictive models over a broad range of selectivity space. Using support vector machines and deep feed-forward neural networks, we demonstrate accurate predictive modeling in the chiral phosphoric acid–catalyzed thiol addition to N-acylimines.",
author = "Zahrt, {Andrew F.} and Henle, {Jeremy J.} and Rose, {Brennan T.} and Yang Wang and Darrow, {William T.} and Denmark, {Scott E.}",
year = "2019",
month = "1",
day = "18",
doi = "10.1126/science.aau5631",
language = "English (US)",
volume = "363",
journal = "Science",
issn = "0036-8075",
publisher = "American Association for the Advancement of Science",
number = "6424",

}

TY - JOUR

T1 - Prediction of higher-selectivity catalysts by computer-driven workflow and machine learning

AU - Zahrt, Andrew F.

AU - Henle, Jeremy J.

AU - Rose, Brennan T.

AU - Wang, Yang

AU - Darrow, William T.

AU - Denmark, Scott E.

PY - 2019/1/18

Y1 - 2019/1/18

N2 - Catalyst design in asymmetric reaction development has traditionally been driven by empiricism, wherein experimentalists attempt to qualitatively recognize structural patterns to improve selectivity. Machine learning algorithms and chemoinformatics can potentially accelerate this process by recognizing otherwise inscrutable patterns in large datasets. Herein we report a computationally guided workflow for chiral catalyst selection using chemoinformatics at every stage of development. Robust molecular descriptors that are agnostic to the catalyst scaffold allow for selection of a universal training set on the basis of steric and electronic properties. This set can be used to train machine learning methods to make highly accurate predictive models over a broad range of selectivity space. Using support vector machines and deep feed-forward neural networks, we demonstrate accurate predictive modeling in the chiral phosphoric acid–catalyzed thiol addition to N-acylimines.

AB - Catalyst design in asymmetric reaction development has traditionally been driven by empiricism, wherein experimentalists attempt to qualitatively recognize structural patterns to improve selectivity. Machine learning algorithms and chemoinformatics can potentially accelerate this process by recognizing otherwise inscrutable patterns in large datasets. Herein we report a computationally guided workflow for chiral catalyst selection using chemoinformatics at every stage of development. Robust molecular descriptors that are agnostic to the catalyst scaffold allow for selection of a universal training set on the basis of steric and electronic properties. This set can be used to train machine learning methods to make highly accurate predictive models over a broad range of selectivity space. Using support vector machines and deep feed-forward neural networks, we demonstrate accurate predictive modeling in the chiral phosphoric acid–catalyzed thiol addition to N-acylimines.

UR - http://www.scopus.com/inward/record.url?scp=85060179885&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060179885&partnerID=8YFLogxK

U2 - 10.1126/science.aau5631

DO - 10.1126/science.aau5631

M3 - Article

C2 - 30655414

AN - SCOPUS:85060179885

VL - 363

JO - Science

JF - Science

SN - 0036-8075

IS - 6424

M1 - eaau5631

ER -