Academic Journal
Προβολή σε EDS

Atomic Uncertainty Pinpoints Critical Failure Structures for Trustworthy Molecular Property Prediction.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Atomic Uncertainty Pinpoints Critical Failure Structures for Trustworthy Molecular Property Prediction.
Συγγραφείς: Qian J; School of Mathematics and Statistics, Central South University, Changsha410083, People's Republic of China., Luo Y; Institute of Interdisciplinary Integrative Medicine Research, Shanghai University of Traditional Chinese Medicine, Shanghai201203, People's Republic of China., Zhou Q; School of Mathematics and Statistics, Central South University, Changsha410083, People's Republic of China., Guan X; Institute of Interdisciplinary Integrative Medicine Research, Shanghai University of Traditional Chinese Medicine, Shanghai201203, People's Republic of China.
Πηγή: Journal of chemical information and modeling [J Chem Inf Model] 2026 Jul 27; Vol. 66 (14), pp. 8076-8086. Date of Electronic Publication: 2026 Jul 14.
Τύπος έκδοσης: Journal Article
Γλώσσα: English
Στοιχεία περιοδικού: Publisher: American Chemical Society Country of Publication: United States NLM ID: 101230060 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1549-960X (Electronic) Linking ISSN: 15499596 NLM ISO Abbreviation: J Chem Inf Model Subsets: MEDLINE
Imprint Name(s): Original Publication: Washington, D.C. : American Chemical Society, c2005-
Ιατρικοί όροι (MeSH): Models, Molecular* , Uncertainty* , Drug Discovery*, Monte Carlo Method ; Solubility
Περίληψη: Accurate prediction of molecular properties is essential for accelerating drug discovery, yet current deep learning methods generally lack reliable uncertainty estimates, particularly at the atomic level. A limitation of current approaches is the reliance on global molecular uncertainty, which frequently masks localized ambiguities by averaging the uncertainty across the entire molecule. To address this, we introduce AUINet, an atomic uncertainty-aware iterative network that quantifies and refines uncertainty at the atomic scale. Built on a D-MPNN architecture, AUINet uses Monte Carlo dropout to estimate atom-level uncertainty and iteratively refines atomic features through uncertainty-guided updates. Comprehensive evaluations show that AUINet outperforms state-of-the-art models on molecular property benchmarks and protein-protein interaction inhibitor tasks under low-data conditions, all without requiring extensive pretraining. Crucially, rejection sampling experiments reveal that atom uncertainty provides a more robust error signal than classic molecular uncertainty. More importantly, AUINet provides chemically interpretable insights by pinpointing specific functional groups and structural motifs that contribute most to prediction uncertainty, as validated in solubility prediction and activity-cliff analysis. Overall, AUINet's precise localization of atomic-level uncertainty establishes a new paradigm for trustworthy molecular property prediction.
Entry Date(s): Date Created: 20260714 Date Completed: 20260805 Latest Revision: 20260805
Update Code: 20260805
DOI: 10.1021/acs.jcim.6c00973
PMID: 42447216
Βάση Δεδομένων: MEDLINE
FullText Text:
  Availability: 0
Header DbId: cmedm
DbLabel: MEDLINE
An: 42447216
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Atomic Uncertainty Pinpoints Critical Failure Structures for Trustworthy Molecular Property Prediction.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AU" term="%22Qian+J%22">Qian J</searchLink>; School of Mathematics and Statistics, Central South University, Changsha410083, People's Republic of China.<br /><searchLink fieldCode="AU" term="%22Luo+Y%22">Luo Y</searchLink>; Institute of Interdisciplinary Integrative Medicine Research, Shanghai University of Traditional Chinese Medicine, Shanghai201203, People's Republic of China.<br /><searchLink fieldCode="AU" term="%22Zhou+Q%22">Zhou Q</searchLink>; School of Mathematics and Statistics, Central South University, Changsha410083, People's Republic of China.<br /><searchLink fieldCode="AU" term="%22Guan+X%22">Guan X</searchLink>; Institute of Interdisciplinary Integrative Medicine Research, Shanghai University of Traditional Chinese Medicine, Shanghai201203, People's Republic of China.
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22101230060%22">Journal of chemical information and modeling</searchLink> [J Chem Inf Model] 2026 Jul 27; Vol. 66 (14), pp. 8076-8086. <i>Date of Electronic Publication: </i>2026 Jul 14.
– Name: TypePub
  Label: Publication Type
  Group: TypPub
  Data: Journal Article
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: TitleSource
  Label: Journal Info
  Group: Src
  Data: <i>Publisher: </i><searchLink fieldCode="PB" term="%22American+Chemical+Society%22">American Chemical Society </searchLink><i>Country of Publication: </i>United States <i>NLM ID: </i>101230060 <i>Publication Model: </i>Print-Electronic <i>Cited Medium: </i>Internet <i>ISSN: </i>1549-960X (Electronic) <i>Linking ISSN: </i><searchLink fieldCode="IS" term="%2215499596%22">15499596 </searchLink><i>NLM ISO Abbreviation: </i>J Chem Inf Model <i>Subsets: </i>MEDLINE
– Name: PublisherInfo
  Label: Imprint Name(s)
  Group: PubInfo
  Data: <i>Original Publication</i>: Washington, D.C. : American Chemical Society, c2005-
– Name: SubjectMESH
  Label: MeSH Terms
  Group: Su
  Data: <searchLink fieldCode="MM" term="%22Models%2C+Molecular%22">Models, Molecular*</searchLink> <br /><searchLink fieldCode="MM" term="%22Uncertainty%22">Uncertainty*</searchLink> <br /><searchLink fieldCode="MM" term="%22Drug+Discovery%22">Drug Discovery*</searchLink><br /><searchLink fieldCode="MH" term="%22Monte+Carlo+Method%22">Monte Carlo Method</searchLink> ; <searchLink fieldCode="MH" term="%22Solubility%22">Solubility</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Accurate prediction of molecular properties is essential for accelerating drug discovery, yet current deep learning methods generally lack reliable uncertainty estimates, particularly at the atomic level. A limitation of current approaches is the reliance on global molecular uncertainty, which frequently masks localized ambiguities by averaging the uncertainty across the entire molecule. To address this, we introduce AUINet, an atomic uncertainty-aware iterative network that quantifies and refines uncertainty at the atomic scale. Built on a D-MPNN architecture, AUINet uses Monte Carlo dropout to estimate atom-level uncertainty and iteratively refines atomic features through uncertainty-guided updates. Comprehensive evaluations show that AUINet outperforms state-of-the-art models on molecular property benchmarks and protein-protein interaction inhibitor tasks under low-data conditions, all without requiring extensive pretraining. Crucially, rejection sampling experiments reveal that atom uncertainty provides a more robust error signal than classic molecular uncertainty. More importantly, AUINet provides chemically interpretable insights by pinpointing specific functional groups and structural motifs that contribute most to prediction uncertainty, as validated in solubility prediction and activity-cliff analysis. Overall, AUINet's precise localization of atomic-level uncertainty establishes a new paradigm for trustworthy molecular property prediction.
– Name: DateEntry
  Label: Entry Date(s)
  Group: Date
  Data: <i>Date Created: </i>20260714 <i>Date Completed: </i>20260805 <i>Latest Revision: </i>20260805
– Name: DateUpdate
  Label: Update Code
  Group: Date
  Data: 20260805
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1021/acs.jcim.6c00973
– Name: AN
  Label: PMID
  Group: ID
  Data: 42447216
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=cmedm&AN=42447216
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1021/acs.jcim.6c00973
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        StartPage: 8076
    Subjects:
      – SubjectFull: Monte Carlo Method
        Type: general
      – SubjectFull: Solubility
        Type: general
      – SubjectFull: Models, Molecular
        Type: general
      – SubjectFull: Uncertainty
        Type: general
      – SubjectFull: Drug Discovery
        Type: general
    Titles:
      – TitleFull: Atomic Uncertainty Pinpoints Critical Failure Structures for Trustworthy Molecular Property Prediction.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Qian J
      – PersonEntity:
          Name:
            NameFull: Luo Y
      – PersonEntity:
          Name:
            NameFull: Zhou Q
      – PersonEntity:
          Name:
            NameFull: Guan X
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 27
              M: 07
              Text: 2026 Jul 27
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-electronic
              Value: 1549-960X
          Numbering:
            – Type: volume
              Value: 66
            – Type: issue
              Value: 14
          Titles:
            – TitleFull: Journal of chemical information and modeling
              Type: main
ResultId 1