Graph Convolutional Neural Networks as "General-Purpose" Property Predictors: The Universality and Limits of Applicability

Nowadays the development of new functional materials/chemical compounds using machine learning (ML) techniques is a hot topic and includes several crucial steps, one of which is the choice of chemical structure representation. The classical approach of rigorous feature engineering in ML typically im...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of chemical information and modeling Ročník 60; číslo 1; s. 22
Hlavní autoři: Korolev, Vadim, Mitrofanov, Artem, Korotcov, Alexandru, Tkachenko, Valery
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States 27.01.2020
Témata:
ISSN:1549-960X, 1549-960X
On-line přístup:Zjistit podrobnosti o přístupu
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Nowadays the development of new functional materials/chemical compounds using machine learning (ML) techniques is a hot topic and includes several crucial steps, one of which is the choice of chemical structure representation. The classical approach of rigorous feature engineering in ML typically improves the performance of the predictive model, but at the same time, it narrows down the scope of applicability and decreases the physical interpretability of predicted results. In this study, we present graph convolutional neural networks (GCNNs) as an architecture that allows for successfully predicting the properties of compounds from diverse domains of chemical space, using a minimal set of meaningful descriptors. The applicability of GCNN models has been demonstrated by a wide range of chemical domain-specific properties. Their performance is comparable to state-of-the-art techniques; however, this architecture exempts from the need to carry out precise feature engineering.
AbstractList Nowadays the development of new functional materials/chemical compounds using machine learning (ML) techniques is a hot topic and includes several crucial steps, one of which is the choice of chemical structure representation. The classical approach of rigorous feature engineering in ML typically improves the performance of the predictive model, but at the same time, it narrows down the scope of applicability and decreases the physical interpretability of predicted results. In this study, we present graph convolutional neural networks (GCNNs) as an architecture that allows for successfully predicting the properties of compounds from diverse domains of chemical space, using a minimal set of meaningful descriptors. The applicability of GCNN models has been demonstrated by a wide range of chemical domain-specific properties. Their performance is comparable to state-of-the-art techniques; however, this architecture exempts from the need to carry out precise feature engineering.
Nowadays the development of new functional materials/chemical compounds using machine learning (ML) techniques is a hot topic and includes several crucial steps, one of which is the choice of chemical structure representation. The classical approach of rigorous feature engineering in ML typically improves the performance of the predictive model, but at the same time, it narrows down the scope of applicability and decreases the physical interpretability of predicted results. In this study, we present graph convolutional neural networks (GCNNs) as an architecture that allows for successfully predicting the properties of compounds from diverse domains of chemical space, using a minimal set of meaningful descriptors. The applicability of GCNN models has been demonstrated by a wide range of chemical domain-specific properties. Their performance is comparable to state-of-the-art techniques; however, this architecture exempts from the need to carry out precise feature engineering.Nowadays the development of new functional materials/chemical compounds using machine learning (ML) techniques is a hot topic and includes several crucial steps, one of which is the choice of chemical structure representation. The classical approach of rigorous feature engineering in ML typically improves the performance of the predictive model, but at the same time, it narrows down the scope of applicability and decreases the physical interpretability of predicted results. In this study, we present graph convolutional neural networks (GCNNs) as an architecture that allows for successfully predicting the properties of compounds from diverse domains of chemical space, using a minimal set of meaningful descriptors. The applicability of GCNN models has been demonstrated by a wide range of chemical domain-specific properties. Their performance is comparable to state-of-the-art techniques; however, this architecture exempts from the need to carry out precise feature engineering.
Author Korolev, Vadim
Mitrofanov, Artem
Korotcov, Alexandru
Tkachenko, Valery
Author_xml – sequence: 1
  givenname: Vadim
  orcidid: 0000-0001-6117-5662
  surname: Korolev
  fullname: Korolev, Vadim
  organization: Department of Chemistry , Lomonosov Moscow State University , Leninskie gory, 1 bld. 3 , Moscow 119991 , Russia
– sequence: 2
  givenname: Artem
  orcidid: 0000-0001-8891-6862
  surname: Mitrofanov
  fullname: Mitrofanov, Artem
  organization: Department of Chemistry , Lomonosov Moscow State University , Leninskie gory, 1 bld. 3 , Moscow 119991 , Russia
– sequence: 3
  givenname: Alexandru
  surname: Korotcov
  fullname: Korotcov, Alexandru
  organization: Science Data Software, LLC , 14909 Forest Landing Circle , Rockville , Maryland 20850 , United States
– sequence: 4
  givenname: Valery
  surname: Tkachenko
  fullname: Tkachenko, Valery
  organization: Science Data Software, LLC , 14909 Forest Landing Circle , Rockville , Maryland 20850 , United States
BackLink https://www.ncbi.nlm.nih.gov/pubmed/31860296$$D View this record in MEDLINE/PubMed
BookMark eNpNUDlPwzAYtVARPWBnQlYnlhQfiZOwVRUUpAo6tBJb5DhfVJfEDnZSVH494ZKY3qk3vDEaGGsAoUtKZpQweiOVn-2VrmdpTkiUxCdoRKMwDVJBXgb_-BCNvd8Twnkq2BkacpoIwlIxQh9LJ5sdXlhzsFXXamtkhZ-gc9_Qvlv36rH0eLoEA70ZrDvXWA9TvHa2AdceewKFVq11_hZvdoC3Rh_AeVnpPpSmwCtd69ZjW-J501RayVx_ZefotJSVh4tfnKDt_d1m8RCsnpePi_kqkCGhbRCSgtKQFgkwELxMykICob2OYsIJcC4iolgoEoBI5SUtgKkYIqEAVMLCmE3Q9c9u4-xbB77Nau0VVJU0YDufMc7SmKWU8L569Vvt8hqKrHG6lu6Y_f3FPgF9NnHu
CitedBy_id crossref_primary_10_1002_anie_202009467
crossref_primary_10_1063_5_0219742
crossref_primary_10_1002_minf_202300327
crossref_primary_10_1038_s41598_025_15265_5
crossref_primary_10_1002_ange_201909989
crossref_primary_10_3390_ijms22169099
crossref_primary_10_1111_jcmm_17652
crossref_primary_10_1155_2021_3833562
crossref_primary_10_1007_s42107_024_01034_8
crossref_primary_10_3390_ma18030534
crossref_primary_10_1016_j_eswa_2023_121753
crossref_primary_10_1039_D1SC02087K
crossref_primary_10_1038_s41598_025_90839_x
crossref_primary_10_1016_j_drudis_2022_03_017
crossref_primary_10_1016_j_mencom_2021_11_003
crossref_primary_10_1038_s41570_021_00260_x
crossref_primary_10_1093_bib_bbac099
crossref_primary_10_1093_bib_bbac131
crossref_primary_10_1088_2515_7639_ad3d89
crossref_primary_10_1515_ntrev_2025_0200
crossref_primary_10_1093_bib_bbab317
crossref_primary_10_1186_s13321_023_00723_x
crossref_primary_10_1002_adma_202413695
crossref_primary_10_1021_acs_macromol_4c02859
crossref_primary_10_1016_j_aca_2022_340558
crossref_primary_10_1016_j_matdes_2022_111162
crossref_primary_10_1021_acsomega_5c00042
crossref_primary_10_1039_D0MH00881H
crossref_primary_10_1007_s12613_022_2547_8
crossref_primary_10_3389_fchem_2021_820417
crossref_primary_10_1016_j_jclepro_2025_145798
crossref_primary_10_1093_bib_bbab112
crossref_primary_10_1016_j_crfs_2025_101127
crossref_primary_10_1016_j_compchemeng_2025_109253
crossref_primary_10_1093_bib_bbac288
crossref_primary_10_1002_aisy_202200243
crossref_primary_10_1039_D1SC01000J
crossref_primary_10_1021_acs_jcim_4c02145
crossref_primary_10_1093_bib_bbab393
crossref_primary_10_1021_acs_jctc_4c01386
crossref_primary_10_3390_ijms23020645
crossref_primary_10_1038_s41598_024_83090_3
crossref_primary_10_1002_aic_17721
crossref_primary_10_1002_ange_202009467
crossref_primary_10_1016_j_micromeso_2021_111666
crossref_primary_10_1186_s13321_020_00479_8
crossref_primary_10_1002_anie_201909989
crossref_primary_10_1139_cjc_2023_0152
crossref_primary_10_1016_j_compchemeng_2023_108392
crossref_primary_10_1002_minf_201900101
crossref_primary_10_1007_s40843_022_2103_9
crossref_primary_10_1038_s43588_024_00699_0
crossref_primary_10_1007_s40820_025_01797_y
crossref_primary_10_1146_annurev_matsci_070218_010015
crossref_primary_10_5668_JEHS_2025_51_2_62
ContentType Journal Article
DBID CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1021/acs.jcim.9b00587
DatabaseName Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE
MEDLINE - Academic
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 no_fulltext_linktorsrc
Discipline Chemistry
EISSN 1549-960X
ExternalDocumentID 31860296
Genre Research Support, Non-U.S. Gov't
Journal Article
GroupedDBID ---
-~X
4.4
55A
5GY
5VS
7~N
AABXI
ABBLG
ABJNI
ABLBI
ABMVS
ABQRX
ABUCX
ACGFS
ACIWK
ACNCT
ACS
ADHLV
AEESW
AENEX
AFEFF
AHGAQ
ALMA_UNASSIGNED_HOLDINGS
AQSVZ
CGR
CUPRZ
CUY
CVF
D0L
DU5
EBS
ECM
ED~
EIF
F5P
GGK
GNL
IH9
JG~
NPM
P2P
PQQKQ
RNS
ROL
UI2
VF5
VG9
W1F
7X8
ID FETCH-LOGICAL-a401t-40d1141d8e2e63f8fdae011d857030e33650c2468ee5cbf1de2c7e56ceec82472
IEDL.DBID 7X8
ISICitedReferencesCount 79
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000510104100005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1549-960X
IngestDate Wed Oct 01 14:58:22 EDT 2025
Mon Jul 21 06:01:44 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 1
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a401t-40d1141d8e2e63f8fdae011d857030e33650c2468ee5cbf1de2c7e56ceec82472
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0001-8891-6862
0000-0001-6117-5662
PMID 31860296
PQID 2329729103
PQPubID 23479
ParticipantIDs proquest_miscellaneous_2329729103
pubmed_primary_31860296
PublicationCentury 2000
PublicationDate 2020-01-27
PublicationDateYYYYMMDD 2020-01-27
PublicationDate_xml – month: 01
  year: 2020
  text: 2020-01-27
  day: 27
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle Journal of chemical information and modeling
PublicationTitleAlternate J Chem Inf Model
PublicationYear 2020
SSID ssj0033962
Score 2.5550134
Snippet Nowadays the development of new functional materials/chemical compounds using machine learning (ML) techniques is a hot topic and includes several crucial...
SourceID proquest
pubmed
SourceType Aggregation Database
Index Database
StartPage 22
SubjectTerms Algorithms
Crystallization
Density Functional Theory
Machine Learning
Models, Molecular
Neural Networks, Computer
Structure-Activity Relationship
Title Graph Convolutional Neural Networks as "General-Purpose" Property Predictors: The Universality and Limits of Applicability
URI https://www.ncbi.nlm.nih.gov/pubmed/31860296
https://www.proquest.com/docview/2329729103
Volume 60
WOSCitedRecordID wos000510104100005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1bS8MwFA7qBH3xfpk34vA1W9v0kvoiYzh9cfRBoW8lS09hgu1c52D-ek_STJ8EwZe20KaE5PR8X3JOz0fITRwhi5eBYOA6kvkCcjZGZGOODBHPQSEFL4zYRDQaiTSNE7vhVtu0ypVPNI46r5TeI-8h8uN7Edz43fSdadUoHV21EhrrpMWRyuiUrij9jiJwHhtBUV2FjCFTT22YEmGtJ1XdfVWTt66pCSii3wmmAZrh7n-7uEd2LMWk_cYm9skalAdka7BSdjsknw-6SjUdVOXCGh4-rqt0mJNJC6-prGnH1qRmCc5GVUOHJnrrfjZf4oUO8GilnluKlkZtfoch9VSWOTU_TtW0Kmi_CZGbJNzlEXkZ3j8PHpnVYGASV15zXF7mOF1uLsCDkBeiyCWgS8h1XXzuAOfI8JTnhwIgUOPCzcFTEQQhYq8Snh95x2SjrEo4JZRzCLCFH8pi7CsRSgFGshRBVOEt3ibXq2HNcEB04EKWUH3U2c_AtslJMzfZtCnGkaFPCh0vDs_-0PqcbHt6uey4zIsuSKvALxwuyaZazCf17MoYDx5HydMXqCvQog
linkProvider ProQuest
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=Graph+Convolutional+Neural+Networks+as+%22General-Purpose%22+Property+Predictors%3A+The+Universality+and+Limits+of+Applicability&rft.jtitle=Journal+of+chemical+information+and+modeling&rft.au=Korolev%2C+Vadim&rft.au=Mitrofanov%2C+Artem&rft.au=Korotcov%2C+Alexandru&rft.au=Tkachenko%2C+Valery&rft.date=2020-01-27&rft.issn=1549-960X&rft.eissn=1549-960X&rft.volume=60&rft.issue=1&rft.spage=22&rft_id=info:doi/10.1021%2Facs.jcim.9b00587&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1549-960X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1549-960X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1549-960X&client=summon