Report
View in EDS

GEM-2: Next Generation Molecular Property Prediction Network by Modeling Full-range Many-body Interactions

Bibliographic Details
Title: GEM-2: Next Generation Molecular Property Prediction Network by Modeling Full-range Many-body Interactions
Authors: Liu, Lihang, He, Donglong, Fang, Xiaomin, Zhang, Shanzhuo, Wang, Fan, He, Jingzhou, Wu, Hua
Publication Year: 2022
Collection: Computer Science
Physics (Other)
Quantitative Biology
Subject Terms: Computer Science - Machine Learning, Physics - Chemical Physics, Quantitative Biology - Molecular Networks, Quantitative Biology - Quantitative Methods
More Details: Molecular property prediction is a fundamental task in the drug and material industries. Physically, the properties of a molecule are determined by its own electronic structure, which is a quantum many-body system and can be exactly described by the Schr"odinger equation. Full-range many-body interactions between electrons have been proven effective in obtaining an accurate solution of the Schr"odinger equation by classical computational chemistry methods, although modeling such interactions consumes an expensive computational cost. Meanwhile, deep learning methods have also demonstrated their competence in molecular property prediction tasks. Inspired by the classical computational chemistry methods, we design a novel method, namely GEM-2, which comprehensively considers full-range many-body interactions in molecules. Multiple tracks are utilized to model the full-range interactions between the many-bodies with different orders, and a novel axial attention mechanism is designed to approximate the full-range interaction modeling with much lower computational cost. Extensive experiments demonstrate the overwhelming superiority of GEM-2 over multiple baseline methods in quantum chemistry and drug discovery tasks. The ablation studies also verify the effectiveness of the full-range many-body interactions.
Document Type: Working Paper
Access URL: http://arxiv.org/abs/2208.05863
Accession Number: edsarx.2208.05863
Database: arXiv
FullText Text:
  Availability: 0
CustomLinks:
  – Url: http://arxiv.org/abs/2208.05863
    Name: EDS - Arxiv
    Category: fullText
    Text: View this record from Arxiv
    MouseOverText: View this record from Arxiv
  – Url: https://resolver.ebsco.com/c/xy5jbn/result?sid=EBSCO:edsarx&genre=article&issn=&ISBN=&volume=&issue=&date=20220811&spage=&pages=&title=GEM-2: Next Generation Molecular Property Prediction Network by Modeling Full-range Many-body Interactions&atitle=GEM-2%3A%20Next%20Generation%20Molecular%20Property%20Prediction%20Network%20by%20Modeling%20Full-range%20Many-body%20Interactions&aulast=Liu%2C%20Lihang&id=DOI:
    Name: Full Text Finder (for New FTF UI) (s8985755)
    Category: fullText
    Text: Find It @ SCU Libraries
    MouseOverText: Find It @ SCU Libraries
Header DbId: edsarx
DbLabel: arXiv
An: edsarx.2208.05863
RelevancyScore: 1037
AccessLevel: 3
PubType: Report
PubTypeId: report
PreciseRelevancyScore: 1037.14245605469
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: GEM-2: Next Generation Molecular Property Prediction Network by Modeling Full-range Many-body Interactions
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Liu%2C+Lihang%22">Liu, Lihang</searchLink><br /><searchLink fieldCode="AR" term="%22He%2C+Donglong%22">He, Donglong</searchLink><br /><searchLink fieldCode="AR" term="%22Fang%2C+Xiaomin%22">Fang, Xiaomin</searchLink><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Shanzhuo%22">Zhang, Shanzhuo</searchLink><br /><searchLink fieldCode="AR" term="%22Wang%2C+Fan%22">Wang, Fan</searchLink><br /><searchLink fieldCode="AR" term="%22He%2C+Jingzhou%22">He, Jingzhou</searchLink><br /><searchLink fieldCode="AR" term="%22Wu%2C+Hua%22">Wu, Hua</searchLink>
– Name: DatePubCY
  Label: Publication Year
  Group: Date
  Data: 2022
– Name: Subset
  Label: Collection
  Group: HoldingsInfo
  Data: Computer Science<br />Physics (Other)<br />Quantitative Biology
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Computer+Science+-+Machine+Learning%22">Computer Science - Machine Learning</searchLink><br /><searchLink fieldCode="DE" term="%22Physics+-+Chemical+Physics%22">Physics - Chemical Physics</searchLink><br /><searchLink fieldCode="DE" term="%22Quantitative+Biology+-+Molecular+Networks%22">Quantitative Biology - Molecular Networks</searchLink><br /><searchLink fieldCode="DE" term="%22Quantitative+Biology+-+Quantitative+Methods%22">Quantitative Biology - Quantitative Methods</searchLink>
– Name: Abstract
  Label: Description
  Group: Ab
  Data: Molecular property prediction is a fundamental task in the drug and material industries. Physically, the properties of a molecule are determined by its own electronic structure, which is a quantum many-body system and can be exactly described by the Schr"odinger equation. Full-range many-body interactions between electrons have been proven effective in obtaining an accurate solution of the Schr"odinger equation by classical computational chemistry methods, although modeling such interactions consumes an expensive computational cost. Meanwhile, deep learning methods have also demonstrated their competence in molecular property prediction tasks. Inspired by the classical computational chemistry methods, we design a novel method, namely GEM-2, which comprehensively considers full-range many-body interactions in molecules. Multiple tracks are utilized to model the full-range interactions between the many-bodies with different orders, and a novel axial attention mechanism is designed to approximate the full-range interaction modeling with much lower computational cost. Extensive experiments demonstrate the overwhelming superiority of GEM-2 over multiple baseline methods in quantum chemistry and drug discovery tasks. The ablation studies also verify the effectiveness of the full-range many-body interactions.
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Working Paper
– Name: URL
  Label: Access URL
  Group: URL
  Data: <link linkTarget="URL" linkTerm="http://arxiv.org/abs/2208.05863" linkWindow="_blank">http://arxiv.org/abs/2208.05863</link>
– Name: AN
  Label: Accession Number
  Group: ID
  Data: edsarx.2208.05863
PLink https://login.libproxy.scu.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edsarx&AN=edsarx.2208.05863
RecordInfo BibRecord:
  BibEntity:
    Subjects:
      – SubjectFull: Computer Science - Machine Learning
        Type: general
      – SubjectFull: Physics - Chemical Physics
        Type: general
      – SubjectFull: Quantitative Biology - Molecular Networks
        Type: general
      – SubjectFull: Quantitative Biology - Quantitative Methods
        Type: general
    Titles:
      – TitleFull: GEM-2: Next Generation Molecular Property Prediction Network by Modeling Full-range Many-body Interactions
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Liu, Lihang
      – PersonEntity:
          Name:
            NameFull: He, Donglong
      – PersonEntity:
          Name:
            NameFull: Fang, Xiaomin
      – PersonEntity:
          Name:
            NameFull: Zhang, Shanzhuo
      – PersonEntity:
          Name:
            NameFull: Wang, Fan
      – PersonEntity:
          Name:
            NameFull: He, Jingzhou
      – PersonEntity:
          Name:
            NameFull: Wu, Hua
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 11
              M: 08
              Type: published
              Y: 2022
ResultId 1