Machine learning and molecular descriptors enable rational solvent selection in asymmetric catalysis

Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Chemical science (Cambridge) Jg. 1; H. 27; S. 6697 - 676
Hauptverfasser: Amar, Yehia, Schweidtmann, Artur M, Deutsch, Paul, Cao, Liwei, Lapkin, Alexei
Format: Journal Article
Sprache:Englisch
Veröffentlicht: England Royal Society of Chemistry 21.07.2019
Schlagworte:
ISSN:2041-6520, 2041-6539
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction-specific descriptors, and additional descriptors based on screening charge density, were calculated. Gaussian process surrogate models were trained on experimental data from a Rh(CO) 2 (acac)/Josiphos catalysed asymmetric hydrogenation of a chiral α-β unsaturated γ-lactam. With two simultaneous objectives - high conversion and high diastereomeric excess - the multi-objective algorithm, trained on the initial dataset of 25 solvents, has identified solvents leading to better reaction outcomes. In addition to being a powerful design of experiments (DoE) methodology, the resulting Gaussian process surrogate model for conversion is, in statistical terms, predictive, with a cross-validation correlation coefficient of 0.84. After identifying promising solvents, the composition of solvent mixtures and optimal reaction temperature were found using a black-box Bayesian optimisation. We then demonstrated the application of a new genetic programming approach to select an appropriate machine learning model for a specific physical system, which should allow the transition of the overall process development workflow into the future robotic laboratories. Rational solvent selection remains a significant challenge in process development.
AbstractList Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction-specific descriptors, and additional descriptors based on screening charge density, were calculated. Gaussian process surrogate models were trained on experimental data from a Rh(CO) 2 (acac)/Josiphos catalysed asymmetric hydrogenation of a chiral α-β unsaturated γ-lactam. With two simultaneous objectives – high conversion and high diastereomeric excess – the multi-objective algorithm, trained on the initial dataset of 25 solvents, has identified solvents leading to better reaction outcomes. In addition to being a powerful design of experiments (DoE) methodology, the resulting Gaussian process surrogate model for conversion is, in statistical terms, predictive, with a cross-validation correlation coefficient of 0.84. After identifying promising solvents, the composition of solvent mixtures and optimal reaction temperature were found using a black-box Bayesian optimisation. We then demonstrated the application of a new genetic programming approach to select an appropriate machine learning model for a specific physical system, which should allow the transition of the overall process development workflow into the future robotic laboratories.
Rational solvent selection remains a significant challenge in process development. Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction-specific descriptors, and additional descriptors based on screening charge density, were calculated. Gaussian process surrogate models were trained on experimental data from a Rh(CO)2(acac)/Josiphos catalysed asymmetric hydrogenation of a chiral α-β unsaturated γ-lactam. With two simultaneous objectives – high conversion and high diastereomeric excess – the multi-objective algorithm, trained on the initial dataset of 25 solvents, has identified solvents leading to better reaction outcomes. In addition to being a powerful design of experiments (DoE) methodology, the resulting Gaussian process surrogate model for conversion is, in statistical terms, predictive, with a cross-validation correlation coefficient of 0.84. After identifying promising solvents, the composition of solvent mixtures and optimal reaction temperature were found using a black-box Bayesian optimisation. We then demonstrated the application of a new genetic programming approach to select an appropriate machine learning model for a specific physical system, which should allow the transition of the overall process development workflow into the future robotic laboratories.
Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction-specific descriptors, and additional descriptors based on screening charge density, were calculated. Gaussian process surrogate models were trained on experimental data from a Rh(CO)2(acac)/Josiphos catalysed asymmetric hydrogenation of a chiral α-β unsaturated γ-lactam. With two simultaneous objectives - high conversion and high diastereomeric excess - the multi-objective algorithm, trained on the initial dataset of 25 solvents, has identified solvents leading to better reaction outcomes. In addition to being a powerful design of experiments (DoE) methodology, the resulting Gaussian process surrogate model for conversion is, in statistical terms, predictive, with a cross-validation correlation coefficient of 0.84. After identifying promising solvents, the composition of solvent mixtures and optimal reaction temperature were found using a black-box Bayesian optimisation. We then demonstrated the application of a new genetic programming approach to select an appropriate machine learning model for a specific physical system, which should allow the transition of the overall process development workflow into the future robotic laboratories.Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction-specific descriptors, and additional descriptors based on screening charge density, were calculated. Gaussian process surrogate models were trained on experimental data from a Rh(CO)2(acac)/Josiphos catalysed asymmetric hydrogenation of a chiral α-β unsaturated γ-lactam. With two simultaneous objectives - high conversion and high diastereomeric excess - the multi-objective algorithm, trained on the initial dataset of 25 solvents, has identified solvents leading to better reaction outcomes. In addition to being a powerful design of experiments (DoE) methodology, the resulting Gaussian process surrogate model for conversion is, in statistical terms, predictive, with a cross-validation correlation coefficient of 0.84. After identifying promising solvents, the composition of solvent mixtures and optimal reaction temperature were found using a black-box Bayesian optimisation. We then demonstrated the application of a new genetic programming approach to select an appropriate machine learning model for a specific physical system, which should allow the transition of the overall process development workflow into the future robotic laboratories.
Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction-specific descriptors, and additional descriptors based on screening charge density, were calculated. Gaussian process surrogate models were trained on experimental data from a Rh(CO) (acac)/Josiphos catalysed asymmetric hydrogenation of a chiral α-β unsaturated γ-lactam. With two simultaneous objectives - high conversion and high diastereomeric excess - the multi-objective algorithm, trained on the initial dataset of 25 solvents, has identified solvents leading to better reaction outcomes. In addition to being a powerful design of experiments (DoE) methodology, the resulting Gaussian process surrogate model for conversion is, in statistical terms, predictive, with a cross-validation correlation coefficient of 0.84. After identifying promising solvents, the composition of solvent mixtures and optimal reaction temperature were found using a black-box Bayesian optimisation. We then demonstrated the application of a new genetic programming approach to select an appropriate machine learning model for a specific physical system, which should allow the transition of the overall process development workflow into the future robotic laboratories.
Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction-specific descriptors, and additional descriptors based on screening charge density, were calculated. Gaussian process surrogate models were trained on experimental data from a Rh(CO) 2 (acac)/Josiphos catalysed asymmetric hydrogenation of a chiral α-β unsaturated γ-lactam. With two simultaneous objectives - high conversion and high diastereomeric excess - the multi-objective algorithm, trained on the initial dataset of 25 solvents, has identified solvents leading to better reaction outcomes. In addition to being a powerful design of experiments (DoE) methodology, the resulting Gaussian process surrogate model for conversion is, in statistical terms, predictive, with a cross-validation correlation coefficient of 0.84. After identifying promising solvents, the composition of solvent mixtures and optimal reaction temperature were found using a black-box Bayesian optimisation. We then demonstrated the application of a new genetic programming approach to select an appropriate machine learning model for a specific physical system, which should allow the transition of the overall process development workflow into the future robotic laboratories. Rational solvent selection remains a significant challenge in process development.
Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared towards automated process development workflow. A library of 459 solvents was used, for which 12 conventional molecular descriptors, two reaction-specific descriptors, and additional descriptors based on screening charge density, were calculated. Gaussian process surrogate models were trained on experimental data from a Rh(CO)2(acac)/Josiphos catalysed asymmetric hydrogenation of a chiral α-β unsaturated γ-lactam. With two simultaneous objectives – high conversion and high diastereomeric excess – the multi-objective algorithm, trained on the initial dataset of 25 solvents, has identified solvents leading to better reaction outcomes. In addition to being a powerful design of experiments (DoE) methodology, the resulting Gaussian process surrogate model for conversion is, in statistical terms, predictive, with a cross-validation correlation coefficient of 0.84. After identifying promising solvents, the composition of solvent mixtures and optimal reaction temperature were found using a black-box Bayesian optimisation. We then demonstrated the application of a new genetic programming approach to select an appropriate machine learning model for a specific physical system, which should allow the transition of the overall process development workflow into the future robotic laboratories.
Author Lapkin, Alexei
Schweidtmann, Artur M
Deutsch, Paul
Amar, Yehia
Cao, Liwei
AuthorAffiliation UCB Pharma S.A. Allée de la Recherche
Aachener Verfahrenstechnik - Process Systems Engineering
Department of Chemical Engineering and Biotechnology
RWTH Aachen University
University of Cambridge
Cambridge Centre for Advanced Research and Education in Singapore Ltd
AuthorAffiliation_xml – sequence: 0
  name: UCB Pharma S.A. Allée de la Recherche
– sequence: 0
  name: Department of Chemical Engineering and Biotechnology
– sequence: 0
  name: Aachener Verfahrenstechnik - Process Systems Engineering
– sequence: 0
  name: Cambridge Centre for Advanced Research and Education in Singapore Ltd
– sequence: 0
  name: RWTH Aachen University
– sequence: 0
  name: University of Cambridge
– name: c UCB Pharma S.A. Allée de la Recherche , 60 1070 , Brussels , Belgium
– name: a Department of Chemical Engineering and Biotechnology , University of Cambridge , Philippa Fawcett Drive , Cambridge , CB3 0AS , UK . Email: aal35@cam.ac.uk
– name: b Aachener Verfahrenstechnik – Process Systems Engineering , RWTH Aachen University , Aachen , Germany
– name: d Cambridge Centre for Advanced Research and Education in Singapore Ltd. , 1 Create Way, CREATE Tower #05-05 , 138602 , Singapore
Author_xml – sequence: 1
  givenname: Yehia
  surname: Amar
  fullname: Amar, Yehia
– sequence: 2
  givenname: Artur M
  surname: Schweidtmann
  fullname: Schweidtmann, Artur M
– sequence: 3
  givenname: Paul
  surname: Deutsch
  fullname: Deutsch, Paul
– sequence: 4
  givenname: Liwei
  surname: Cao
  fullname: Cao, Liwei
– sequence: 5
  givenname: Alexei
  surname: Lapkin
  fullname: Lapkin, Alexei
BackLink https://www.ncbi.nlm.nih.gov/pubmed/31367324$$D View this record in MEDLINE/PubMed
BookMark eNptks1rFTEUxYNUbK3duFcCbkR4mq_JTDaF8vALKi7UdbiTudOmZJJnMlN4_715ffWpxWwScn_3cHJPnpKjmCIS8pyzt5xJ886Z4hjvlIJH5EQwxVe6kebocBbsmJyVcsPqkpI3on1CjiWXupVCnZDhC7hrH5EGhBx9vKIQBzqlgG4JkOmAxWW_mVMuFCP0AWmG2acIgZYUbjHOtGCld3fURwplO004Z--ogxnCtvjyjDweIRQ8u99PyY8P77-vP60uv378vL64XDll1LzCtjMwyL7TvQQp5SgGZA0fXdePI3a8x0YY0RslnVDcsNZI7DoYHTZydD3IU3K-190s_YSDq-YyBLvJfoK8tQm8_bcS_bW9SrdWa9EoI6rA63uBnH4uWGY7-eIwBIiYlmKF0G2rtBRNRV89QG_SkutYdlSjGq0V05V6-bejg5XfAVTgzR5wOZWScTwgnNldwHZtvq3vAr6oMHsAOz_fpVFf48P_W17sW3JxB-k_f0b-AtSrs_I
CitedBy_id crossref_primary_10_1002_cmtd_202000044
crossref_primary_10_1039_D3RE00476G
crossref_primary_10_1039_D2QO01680J
crossref_primary_10_1002_cssc_202301639
crossref_primary_10_1126_science_abj0999
crossref_primary_10_3390_molecules28010322
crossref_primary_10_1016_j_ces_2025_122628
crossref_primary_10_1038_s41467_025_62825_4
crossref_primary_10_1246_bcsj_20200045
crossref_primary_10_1016_j_molliq_2024_125582
crossref_primary_10_1002_chem_202401626
crossref_primary_10_1038_s41529_024_00435_z
crossref_primary_10_1016_j_coche_2021_100752
crossref_primary_10_3390_futurepharmacol5020024
crossref_primary_10_1002_cmtd_202100053
crossref_primary_10_1016_j_jddst_2023_104751
crossref_primary_10_1021_acs_joc_5c01229
crossref_primary_10_1021_acsengineeringau_5c00004
crossref_primary_10_1038_s41929_024_01150_3
crossref_primary_10_1002_wcms_1540
crossref_primary_10_1039_D1RE00549A
crossref_primary_10_1039_D1SC00482D
crossref_primary_10_1002_advs_202105547
crossref_primary_10_1039_D2SC05974F
crossref_primary_10_1093_bib_bbaf286
crossref_primary_10_1016_j_ces_2021_116938
crossref_primary_10_1016_j_memsci_2020_117860
crossref_primary_10_1002_qre_3025
crossref_primary_10_1021_cbe_4c00170
crossref_primary_10_1016_j_trechm_2020_12_001
crossref_primary_10_1038_s41598_022_08997_1
crossref_primary_10_1016_j_cej_2019_123340
crossref_primary_10_1016_j_chemosphere_2022_136447
crossref_primary_10_1039_D4RE00175C
crossref_primary_10_3390_pr11020330
crossref_primary_10_1186_s13321_025_01027_y
crossref_primary_10_1002_aic_18182
crossref_primary_10_1021_acscentsci_3c00050
crossref_primary_10_1070_RCR5012
crossref_primary_10_1007_s12532_021_00204_y
crossref_primary_10_1039_D2QO00448H
crossref_primary_10_1002_cite_202100083
crossref_primary_10_1016_j_fuel_2022_125340
crossref_primary_10_1039_D3RE00628J
crossref_primary_10_1016_j_pmatsci_2022_101043
crossref_primary_10_1002_wcms_1603
crossref_primary_10_1016_j_ces_2024_119881
crossref_primary_10_1039_D3RE00539A
crossref_primary_10_1146_annurev_chembioeng_092320_120230
crossref_primary_10_1039_D2RE00008C
crossref_primary_10_1002_aic_18698
crossref_primary_10_1002_aic_17248
crossref_primary_10_1093_bulcsj_uoaf022
crossref_primary_10_1039_D3CY00083D
crossref_primary_10_1039_D3SC01261A
crossref_primary_10_1016_j_cej_2022_138443
crossref_primary_10_1016_j_mcat_2021_111876
crossref_primary_10_1021_acs_oprd_4c00533
crossref_primary_10_1016_j_cej_2022_139099
crossref_primary_10_1016_j_molliq_2023_122355
crossref_primary_10_3390_suschem3010009
crossref_primary_10_1002_chem_202402207
crossref_primary_10_1016_j_compchemeng_2025_109236
crossref_primary_10_1016_j_ddtec_2020_06_002
crossref_primary_10_1039_D0EE02543G
crossref_primary_10_1016_j_cep_2022_109148
Cites_doi 10.1021/acs.accounts.6b00261
10.1073/pnas.1013331108
10.1016/j.ces.2013.11.020
10.1021/ja01131a010
10.1039/C5CY02197A
10.3762/bjoc.13.18
10.1016/j.ijggc.2016.03.022
10.1186/s13065-015-0104-5
10.1039/C7SC04679K
10.1021/acs.accounts.6b00194
10.1039/b617536h
10.1016/j.compchemeng.2004.08.010
10.3891/acta.chem.scand.39b-0079
10.1002/anie.201705721
10.1038/nchem.1755
10.1016/j.memsci.2018.10.013
10.1007/978-1-84800-382-8
10.1111/j.1528-1167.2012.03598.x
10.1021/j100007a062
10.1039/C6RE00059B
10.1021/op300275p
10.1016/j.ces.2018.12.003
10.1039/C8RE00032H
10.1016/j.cej.2018.07.031
10.1039/C5CC03651H
10.1021/acssuschemeng.5b01721
10.1039/b211894g
10.1002/cssc.201700927
10.1093/oso/9780199296590.001.0001
10.1007/s10898-018-0609-2
10.1038/nchem.2501
10.1002/aic.14630
10.1021/jp050395e
10.1126/science.aar5169
10.1002/1521-3773(20020415)41:8<1335::AID-ANIE1335>3.0.CO;2-A
10.1039/b921393g
10.1021/om100648v
10.1016/j.ces.2015.07.010
10.7551/mitpress/3206.001.0001
10.1021/ja909853x
10.1016/j.compchemeng.2018.07.015
10.1002/anie.201501618
10.1021/ja00287a001
10.1186/s13321-018-0277-8
10.1021/jo048163j
10.1146/annurev.anchem.111808.073718
10.1021/bp00008a006
10.1093/bioinformatics/btm344
10.1038/nrd3078
10.1016/j.tet.2018.02.061
10.1093/biomet/25.3-4.285
10.1039/C5RE00083A
10.1039/C5OB01892G
10.1002/chem.200900317
ContentType Journal Article
Copyright Copyright Royal Society of Chemistry 2019
This journal is © The Royal Society of Chemistry 2019 2019
Copyright_xml – notice: Copyright Royal Society of Chemistry 2019
– notice: This journal is © The Royal Society of Chemistry 2019 2019
DBID AAYXX
CITATION
NPM
7SR
8BQ
8FD
JG9
7X8
5PM
DOI 10.1039/c9sc01844a
DatabaseName CrossRef
PubMed
Engineered Materials Abstracts
METADEX
Technology Research Database
Materials Research Database
MEDLINE - Academic
PubMed Central (Full Participant titles)
DatabaseTitle CrossRef
PubMed
Materials Research Database
Engineered Materials Abstracts
Technology Research Database
METADEX
MEDLINE - Academic
DatabaseTitleList CrossRef

MEDLINE - Academic
PubMed

Materials Research Database
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Chemistry
EISSN 2041-6539
EndPage 676
ExternalDocumentID PMC6625492
31367324
10_1039_C9SC01844A
c9sc01844a
Genre Journal Article
GroupedDBID -JG
0-7
0R~
53G
705
7~J
AAEMU
AAFWJ
AAIWI
AAJAE
AARTK
AAXHV
ABEMK
ABPDG
ABXOH
ACGFS
ACIWK
ADBBV
ADMRA
AEFDR
AENEX
AESAV
AFLYV
AGEGJ
AGRSR
AGSTE
AHGCF
AKBGW
ALMA_UNASSIGNED_HOLDINGS
ANUXI
AOIJS
APEMP
AUDPV
AZFZN
BCNDV
BLAPV
BSQNT
C6K
D0L
EE0
EF-
F5P
GROUPED_DOAJ
H13
HYE
HZ~
H~N
O-G
O9-
OK1
PGMZT
R7C
R7D
RAOCF
RCNCU
RNS
RPM
RRC
RSCEA
RVUXY
SKA
SKF
SKH
SKJ
SKM
SKR
SKZ
SLC
SLF
SLH
SMJ
AAYXX
ABIQK
AFPKN
AGMRB
CITATION
NPM
7SR
8BQ
8FD
JG9
7X8
5PM
ID FETCH-LOGICAL-c494t-e789ad3b86b3a333f2de051fc8bffe81be5292b943c24190793e88afce53fcba3
IEDL.DBID RRC
ISICitedReferencesCount 103
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000477947800012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2041-6520
IngestDate Tue Nov 04 01:58:57 EST 2025
Wed Oct 01 13:58:43 EDT 2025
Fri Jul 25 03:11:13 EDT 2025
Mon Mar 03 18:06:50 EST 2025
Tue Nov 18 21:06:42 EST 2025
Sat Nov 29 05:54:04 EST 2025
Tue Dec 17 21:00:00 EST 2024
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 27
Language English
License This article is freely available. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence (CC BY 3.0)
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c494t-e789ad3b86b3a333f2de051fc8bffe81be5292b943c24190793e88afce53fcba3
Notes 10.1039/c9sc01844a
Electronic supplementary information (ESI) available: Library of solvents and molecular descriptors. See DOI
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ORCID 0000-0001-7621-0889
0000-0003-3097-3762
0000-0001-8885-6847
OpenAccessLink http://dx.doi.org/10.1039/c9sc01844a
PMID 31367324
PQID 2254566406
PQPubID 2047492
PageCount 1
ParticipantIDs crossref_primary_10_1039_C9SC01844A
rsc_primary_c9sc01844a
crossref_citationtrail_10_1039_C9SC01844A
proquest_journals_2254566406
proquest_miscellaneous_2267746325
pubmed_primary_31367324
pubmedcentral_primary_oai_pubmedcentral_nih_gov_6625492
PublicationCentury 2000
PublicationDate 2019-07-21
PublicationDateYYYYMMDD 2019-07-21
PublicationDate_xml – month: 07
  year: 2019
  text: 2019-07-21
  day: 21
PublicationDecade 2010
PublicationPlace England
PublicationPlace_xml – name: England
– name: Cambridge
PublicationTitle Chemical science (Cambridge)
PublicationTitleAlternate Chem Sci
PublicationYear 2019
Publisher Royal Society of Chemistry
Publisher_xml – name: Royal Society of Chemistry
References Todeschini (C9SC01844A-(cit16)/*[position()=1]) 2010
Reizman (C9SC01844A-(cit14)/*[position()=1]) 2016; 49
Taft (C9SC01844A-(cit17)/*[position()=1]) 1952; 74
Zhou (C9SC01844A-(cit43)/*[position()=1]) 2015; 137
Ley (C9SC01844A-(cit8)/*[position()=1]) 2015; 54
Reetz (C9SC01844A-(cit51)/*[position()=1]) 2002; 41
Echtermeyer (C9SC01844A-(cit4)/*[position()=1]) 2017; 13
Stenutz (C9SC01844A-(cit45)/*[position()=1])
Guyon (C9SC01844A-(cit29)/*[position()=1]) 2003; 3
Jessop (C9SC01844A-(cit60)/*[position()=1]) 2003; 5
Murray (C9SC01844A-(cit25)/*[position()=1]) 2016; 14
Capello (C9SC01844A-(cit56)/*[position()=1]) 2007; 9
Awonaike (C9SC01844A-(cit37)/*[position()=1]) 2017; 19
Kreutz (C9SC01844A-(cit13)/*[position()=1]) 2010; 132
Baumgartner (C9SC01844A-(cit11)/*[position()=1]) 2018; 3
Rall (C9SC01844A-(cit32)/*[position()=1]) 2019; 569
McMullen (C9SC01844A-(cit3)/*[position()=1]) 2010; 3
Bradford (C9SC01844A-(cit49)/*[position()=1]) 2018; 118
Fey (C9SC01844A-(cit19)/*[position()=1]) 2015; 9
Jeraal (C9SC01844A-(cit7)/*[position()=1]) 2018; 74
Paul (C9SC01844A-(cit1)/*[position()=1]) 2010; 9
Inza (C9SC01844A-(cit30)/*[position()=1]) 2007; 23
Olson (C9SC01844A-(cit53)/*[position()=1]) 2016; vol. 64
Thompson (C9SC01844A-(cit50)/*[position()=1]) 1933; 25
Martha Cornelius (C9SC01844A-(cit52)/*[position()=1]) 2009; 15
Jover (C9SC01844A-(cit18)/*[position()=1]) 2010; 29
Niemeyer (C9SC01844A-(cit22)/*[position()=1]) 2016; 8
Holmes (C9SC01844A-(cit6)/*[position()=1]) 2016; 1
Van Paesschen (C9SC01844A-(cit33)/*[position()=1]) 2012; 54
Grethe (C9SC01844A-(cit35)/*[position()=1]) 2018; 10
Gani (C9SC01844A-(cit61)/*[position()=1]) 2004; 28
Bentley (C9SC01844A-(cit58)/*[position()=1]) 2005; 70
Murray (C9SC01844A-(cit12)/*[position()=1]) 2013; 17
Atkinson (C9SC01844A-(cit2)/*[position()=1]) 2007
Schweidtmann (C9SC01844A-(cit9)/*[position()=1]) 2018; 352
Holmes (C9SC01844A-(cit5)/*[position()=1]) 2016; 1
Maldonado (C9SC01844A-(cit55)/*[position()=1]) 2010; 39
Klamt (C9SC01844A-(cit40)/*[position()=1]) 1995; 99
Dyson (C9SC01844A-(cit59)/*[position()=1]) 2016; 6
Reizman (C9SC01844A-(cit24)/*[position()=1]) 2015; 51
Rasmussen (C9SC01844A-(cit54)/*[position()=1]) 2005
Duros (C9SC01844A-(cit44)/*[position()=1]) 2017; 56
Bruce (C9SC01844A-(cit57)/*[position()=1]) 1991; 7
Santiago (C9SC01844A-(cit20)/*[position()=1]) 2018; 9
Chastrette (C9SC01844A-(cit26)/*[position()=1]) 1985; 107
Caramia (C9SC01844A-(cit48)/*[position()=1]) 2008
Sigman (C9SC01844A-(cit21)/*[position()=1]) 2016; 49
Bradford (C9SC01844A-(cit46)/*[position()=1]) 2018; 71
Struebing (C9SC01844A-(cit31)/*[position()=1]) 2013; 5
Zhou (C9SC01844A-(cit28)/*[position()=1]) 2014; 115
Zhou (C9SC01844A-(cit42)/*[position()=1]) 2015; 61
Ahneman (C9SC01844A-(cit23)/*[position()=1]) 2018; 360
Helmdach (C9SC01844A-(cit47)/*[position()=1]) 2017; 10
Harper (C9SC01844A-(cit10)/*[position()=1]) 2011; 108
Carlson (C9SC01844A-(cit27)/*[position()=1]) 1985; 38
Turchi (C9SC01844A-(cit38)/*[position()=1]) 2019; 197
Katritzky (C9SC01844A-(cit15)/*[position()=1]) 2005; 109
Suberu (C9SC01844A-(cit36)/*[position()=1]) 2016; 4
Gonfa (C9SC01844A-(cit41)/*[position()=1]) 2016; 49
References_xml – issn: 2010
  publication-title: Molecular Descriptors for Chemoinformatics
  doi: Todeschini Consonni
– publication-title: Dielectric constants and refractive index
  doi: Stenutz
– issn: 2008
  end-page: p 11-36
  publication-title: Multi-objective Management in Freight Logistics: Increasing Capacity, Service Level and Safety with Optimization Algorithms
  doi: Caramia Dell'Olmo
– issn: 2007
  publication-title: Optimum Experimental Design with SAS
  doi: Atkinson Donev Tobias
– publication-title: COSMOtherm, Version C3.0 Release 14.01
– issn: 2016
  issue: vol. 64
  end-page: p 66-74
  publication-title: Proceedings of the Workshop on Automatic Machine Learning
  doi: Olson Bartley Urbanowicz Moore
– issn: 2005
  publication-title: Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning)
  doi: Rasmussen Williams
– volume: 49
  start-page: 1786
  year: 2016
  ident: C9SC01844A-(cit14)/*[position()=1]
  publication-title: Acc. Chem. Res.
  doi: 10.1021/acs.accounts.6b00261
– volume: 108
  start-page: 2179
  year: 2011
  ident: C9SC01844A-(cit10)/*[position()=1]
  publication-title: Proc. Natl. Acad. Sci. U. S. A.
  doi: 10.1073/pnas.1013331108
– volume: 115
  start-page: 177
  year: 2014
  ident: C9SC01844A-(cit28)/*[position()=1]
  publication-title: Chem. Eng. Sci.
  doi: 10.1016/j.ces.2013.11.020
– volume: 74
  start-page: 2729
  year: 1952
  ident: C9SC01844A-(cit17)/*[position()=1]
  publication-title: J. Am. Chem. Soc.
  doi: 10.1021/ja01131a010
– volume: 6
  start-page: 3302
  year: 2016
  ident: C9SC01844A-(cit59)/*[position()=1]
  publication-title: Catal. Sci. Technol.
  doi: 10.1039/C5CY02197A
– volume: 13
  start-page: 150
  year: 2017
  ident: C9SC01844A-(cit4)/*[position()=1]
  publication-title: Beilstein J. Org. Chem.
  doi: 10.3762/bjoc.13.18
– volume: 49
  start-page: 372
  year: 2016
  ident: C9SC01844A-(cit41)/*[position()=1]
  publication-title: Int. J. Greenhouse Gas Control
  doi: 10.1016/j.ijggc.2016.03.022
– volume: 9
  start-page: 38
  year: 2015
  ident: C9SC01844A-(cit19)/*[position()=1]
  publication-title: Chem. Cent. J.
  doi: 10.1186/s13065-015-0104-5
– volume: 9
  start-page: 2398
  year: 2018
  ident: C9SC01844A-(cit20)/*[position()=1]
  publication-title: Chem. Sci.
  doi: 10.1039/C7SC04679K
– volume: 49
  start-page: 1292
  year: 2016
  ident: C9SC01844A-(cit21)/*[position()=1]
  publication-title: Acc. Chem. Res.
  doi: 10.1021/acs.accounts.6b00194
– volume: 9
  start-page: 927
  year: 2007
  ident: C9SC01844A-(cit56)/*[position()=1]
  publication-title: Green Chem.
  doi: 10.1039/b617536h
– volume: 28
  start-page: 2441
  year: 2004
  ident: C9SC01844A-(cit61)/*[position()=1]
  publication-title: Comput. Chem. Eng.
  doi: 10.1016/j.compchemeng.2004.08.010
– volume: 38
  start-page: 79
  year: 1985
  ident: C9SC01844A-(cit27)/*[position()=1]
  publication-title: Acta Chem. Scand., Ser. B
  doi: 10.3891/acta.chem.scand.39b-0079
– volume: 56
  start-page: 10815
  year: 2017
  ident: C9SC01844A-(cit44)/*[position()=1]
  publication-title: Angew. Chem., Int. Ed.
  doi: 10.1002/anie.201705721
– volume: 5
  start-page: 952
  year: 2013
  ident: C9SC01844A-(cit31)/*[position()=1]
  publication-title: Nat. Chem.
  doi: 10.1038/nchem.1755
– volume: 569
  start-page: 209
  year: 2019
  ident: C9SC01844A-(cit32)/*[position()=1]
  publication-title: J. Membr. Sci.
  doi: 10.1016/j.memsci.2018.10.013
– volume-title: Multi-objective Management in Freight Logistics: Increasing Capacity, Service Level and Safety with Optimization Algorithms
  year: 2008
  ident: C9SC01844A-(cit48)/*[position()=1]
  doi: 10.1007/978-1-84800-382-8
– volume: 54
  start-page: 89
  year: 2012
  ident: C9SC01844A-(cit33)/*[position()=1]
  publication-title: Epilepsia
  doi: 10.1111/j.1528-1167.2012.03598.x
– volume: 99
  start-page: 2224
  year: 1995
  ident: C9SC01844A-(cit40)/*[position()=1]
  publication-title: J. Phys. Chem.
  doi: 10.1021/j100007a062
– volume: 1
  start-page: 366
  year: 2016
  ident: C9SC01844A-(cit6)/*[position()=1]
  publication-title: React. Chem. Eng.
  doi: 10.1039/C6RE00059B
– volume: 17
  start-page: 40
  year: 2013
  ident: C9SC01844A-(cit12)/*[position()=1]
  publication-title: Org. Process Res. Dev.
  doi: 10.1021/op300275p
– volume: 197
  start-page: 150
  year: 2019
  ident: C9SC01844A-(cit38)/*[position()=1]
  publication-title: Chem. Eng. Sci.
  doi: 10.1016/j.ces.2018.12.003
– volume: 3
  start-page: 301
  year: 2018
  ident: C9SC01844A-(cit11)/*[position()=1]
  publication-title: React. Chem. Eng.
  doi: 10.1039/C8RE00032H
– volume: 352
  start-page: 277
  year: 2018
  ident: C9SC01844A-(cit9)/*[position()=1]
  publication-title: Chem. Eng. J.
  doi: 10.1016/j.cej.2018.07.031
– volume: 51
  start-page: 13290
  year: 2015
  ident: C9SC01844A-(cit24)/*[position()=1]
  publication-title: Chem. Commun.
  doi: 10.1039/C5CC03651H
– volume: 4
  start-page: 2559
  year: 2016
  ident: C9SC01844A-(cit36)/*[position()=1]
  publication-title: ACS Sustainable Chem. Eng.
  doi: 10.1021/acssuschemeng.5b01721
– volume: 5
  start-page: 123
  year: 2003
  ident: C9SC01844A-(cit60)/*[position()=1]
  publication-title: Green Chem.
  doi: 10.1039/b211894g
– volume: 10
  start-page: 3632
  year: 2017
  ident: C9SC01844A-(cit47)/*[position()=1]
  publication-title: ChemSusChem
  doi: 10.1002/cssc.201700927
– volume-title: Optimum Experimental Design with SAS
  year: 2007
  ident: C9SC01844A-(cit2)/*[position()=1]
  doi: 10.1093/oso/9780199296590.001.0001
– volume: 19
  start-page: 288
  year: 2017
  ident: C9SC01844A-(cit37)/*[position()=1]
  publication-title: Environ. Sci.: Processes Impacts
– volume: 71
  start-page: 407
  year: 2018
  ident: C9SC01844A-(cit46)/*[position()=1]
  publication-title: J. Glob. Optim.
  doi: 10.1007/s10898-018-0609-2
– volume-title: Molecular Descriptors for Chemoinformatics
  year: 2010
  ident: C9SC01844A-(cit16)/*[position()=1]
– volume: 8
  start-page: 610
  year: 2016
  ident: C9SC01844A-(cit22)/*[position()=1]
  publication-title: Nat. Chem.
  doi: 10.1038/nchem.2501
– volume-title: Dielectric constants and refractive index
  ident: C9SC01844A-(cit45)/*[position()=1]
– volume: 61
  start-page: 147
  year: 2015
  ident: C9SC01844A-(cit42)/*[position()=1]
  publication-title: AIChE J.
  doi: 10.1002/aic.14630
– volume: 109
  start-page: 10323
  year: 2005
  ident: C9SC01844A-(cit15)/*[position()=1]
  publication-title: J. Phys. Chem. A
  doi: 10.1021/jp050395e
– volume: 360
  start-page: 186
  year: 2018
  ident: C9SC01844A-(cit23)/*[position()=1]
  publication-title: Science
  doi: 10.1126/science.aar5169
– volume: 41
  start-page: 1335
  year: 2002
  ident: C9SC01844A-(cit51)/*[position()=1]
  publication-title: Angew. Chem., Int. Ed.
  doi: 10.1002/1521-3773(20020415)41:8<1335::AID-ANIE1335>3.0.CO;2-A
– volume: 39
  start-page: 1891
  year: 2010
  ident: C9SC01844A-(cit55)/*[position()=1]
  publication-title: Chem. Soc. Rev.
  doi: 10.1039/b921393g
– volume: 29
  start-page: 6245
  year: 2010
  ident: C9SC01844A-(cit18)/*[position()=1]
  publication-title: Organometallics
  doi: 10.1021/om100648v
– volume: 137
  start-page: 613
  year: 2015
  ident: C9SC01844A-(cit43)/*[position()=1]
  publication-title: Chem. Eng. Sci.
  doi: 10.1016/j.ces.2015.07.010
– volume-title: Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning)
  year: 2005
  ident: C9SC01844A-(cit54)/*[position()=1]
  doi: 10.7551/mitpress/3206.001.0001
– volume: 132
  start-page: 3128
  year: 2010
  ident: C9SC01844A-(cit13)/*[position()=1]
  publication-title: J. Am. Chem. Soc.
  doi: 10.1021/ja909853x
– volume: 118
  start-page: 143
  year: 2018
  ident: C9SC01844A-(cit49)/*[position()=1]
  publication-title: Comput. Chem. Eng.
  doi: 10.1016/j.compchemeng.2018.07.015
– volume: 54
  start-page: 10122
  year: 2015
  ident: C9SC01844A-(cit8)/*[position()=1]
  publication-title: Angew. Chem., Int. Ed.
  doi: 10.1002/anie.201501618
– volume: 107
  start-page: 1
  year: 1985
  ident: C9SC01844A-(cit26)/*[position()=1]
  publication-title: J. Am. Chem. Soc.
  doi: 10.1021/ja00287a001
– volume: vol. 64
  volume-title: Proceedings of the Workshop on Automatic Machine Learning
  year: 2016
  ident: C9SC01844A-(cit53)/*[position()=1]
– volume: 3
  start-page: 1157
  year: 2003
  ident: C9SC01844A-(cit29)/*[position()=1]
  publication-title: J. Mach. Learn. Res.
– volume: 10
  start-page: 22
  year: 2018
  ident: C9SC01844A-(cit35)/*[position()=1]
  publication-title: J. Cheminf.
  doi: 10.1186/s13321-018-0277-8
– volume: 70
  start-page: 1647
  year: 2005
  ident: C9SC01844A-(cit58)/*[position()=1]
  publication-title: J. Org. Chem.
  doi: 10.1021/jo048163j
– volume: 3
  start-page: 19
  year: 2010
  ident: C9SC01844A-(cit3)/*[position()=1]
  publication-title: Annu. Rev. Anal. Chem.
  doi: 10.1146/annurev.anchem.111808.073718
– volume: 7
  start-page: 116
  year: 1991
  ident: C9SC01844A-(cit57)/*[position()=1]
  publication-title: Biotechnol. Prog.
  doi: 10.1021/bp00008a006
– volume: 23
  start-page: 2507
  year: 2007
  ident: C9SC01844A-(cit30)/*[position()=1]
  publication-title: Bioinformatics
  doi: 10.1093/bioinformatics/btm344
– volume: 9
  start-page: 203
  year: 2010
  ident: C9SC01844A-(cit1)/*[position()=1]
  publication-title: Nat. Rev. Drug Discovery
  doi: 10.1038/nrd3078
– volume: 74
  start-page: 3158
  year: 2018
  ident: C9SC01844A-(cit7)/*[position()=1]
  publication-title: Tetrahedron
  doi: 10.1016/j.tet.2018.02.061
– volume: 25
  start-page: 285
  year: 1933
  ident: C9SC01844A-(cit50)/*[position()=1]
  publication-title: Biometrika
  doi: 10.1093/biomet/25.3-4.285
– volume: 1
  start-page: 96
  year: 2016
  ident: C9SC01844A-(cit5)/*[position()=1]
  publication-title: React. Chem. Eng.
  doi: 10.1039/C5RE00083A
– volume: 14
  start-page: 2373
  year: 2016
  ident: C9SC01844A-(cit25)/*[position()=1]
  publication-title: Org. Biomol. Chem.
  doi: 10.1039/C5OB01892G
– volume: 15
  start-page: 7368
  year: 2009
  ident: C9SC01844A-(cit52)/*[position()=1]
  publication-title: Chem.–Eur. J.
  doi: 10.1002/chem.200900317
SSID ssj0000331527
Score 2.5636384
Snippet Rational solvent selection remains a significant challenge in process development. Here we describe a hybrid mechanistic-machine learning approach, geared...
Rational solvent selection remains a significant challenge in process development. Rational solvent selection remains a significant challenge in process...
SourceID pubmedcentral
proquest
pubmed
crossref
rsc
SourceType Open Access Repository
Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 6697
SubjectTerms Amides
Artificial intelligence
Bayesian analysis
Catalysis
Charge density
Chemistry
Conversion
Correlation coefficients
Design of experiments
Gaussian process
Genetic algorithms
Machine learning
Multiple objective analysis
Optimization
Solvents
Workflow
Title Machine learning and molecular descriptors enable rational solvent selection in asymmetric catalysis
URI https://www.ncbi.nlm.nih.gov/pubmed/31367324
https://www.proquest.com/docview/2254566406
https://www.proquest.com/docview/2267746325
https://pubmed.ncbi.nlm.nih.gov/PMC6625492
Volume 1
WOSCitedRecordID wos000477947800012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 2041-6539
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000331527
  issn: 2041-6520
  databaseCode: DOA
  dateStart: 20150101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVAUL
  databaseName: Royal Society Of Chemistry Journals 2008-
  customDbUrl:
  eissn: 2041-6539
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000331527
  issn: 2041-6520
  databaseCode: RRC
  dateStart: 20100101
  isFulltext: true
  titleUrlDefault: https://pubs.rsc.org/
  providerName: Royal Society of Chemistry
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1bS-QwFD6oCO6Lt13Xehki64sPYdskTZtHKYoviniBeStpero7sNORdhT89yaZtjK4gq_NKU2Tk5wvycn3AZzasGJKlUTUsBip0GFBCxGWFIURLqWyipT2YhPJzU06HqvbFTj95ASfq9-Zus9Cuw4R526iTWLnund32bCREnLeSbOyUERUxizsaUiX3l4OPB_Q5MekyNWm1wDxseZy62u13IbNDkuS80Xn78AK1ruwkfUSbt-hvPapkkg6bYg_RNclmfaKuKTExaQxa1qC_hIVabrNQWJ90qVCktYL5dhnZFIT3b5Op06DyxC_7-PoTH7A4-XFQ3ZFO1kFaoQSc4pJqnTJi1QWXHPOK1aiHZqVSYuqQgtjMWaKFUpwY8O7cgx6mKa6MhjzyhSa78FaPatxH4hWRogisasmRBGLUkcKFUOUWsSYyCiAs77Nc9Nxjjvpi3-5P_vmKn9vtwB-DbZPC6aN_1od9V2Xd6Otze2cZHGgtNgkgJOh2La1O_zQNc6enY20SFdyFgfwc9HTw2e4462zyDKAZMkHBgPHwb1cUk_-ei5uKd0KmwWwZ71lsDeqNb6--uBLf3UI3ywQc7fJKIuOYG3ePOMxrJuX-aRtRrCajNOR3y0Yed9_A_jc_KM
linkProvider Royal Society of Chemistry
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Machine+learning+and+molecular+descriptors+enable+rational+solvent+selection+in+asymmetric+catalysis&rft.jtitle=Chemical+science+%28Cambridge%29&rft.au=Amar%2C+Yehia&rft.au=Schweidtmann%2C+Artur%C2%A0M.&rft.au=Deutsch%2C+Paul&rft.au=Cao%2C+Liwei&rft.date=2019-07-21&rft.issn=2041-6520&rft.eissn=2041-6539&rft.volume=10&rft.issue=27&rft.spage=6697&rft.epage=6706&rft_id=info:doi/10.1039%2FC9SC01844A&rft.externalDBID=n%2Fa&rft.externalDocID=10_1039_C9SC01844A
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2041-6520&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2041-6520&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2041-6520&client=summon