Title:
MESSAGE PASSING GRAPH NEURAL NETWORK WITH VECTOR-SCALAR MESSAGE PASSING AND RUN-TIME GEOMETRIC COMPUTATION
Document Type and Number:
WIPO Patent Application WO/2024/082306
Kind Code:
A1
Abstract:
A computing system is provided, which receives a molecular graph at a message passing graph neural network (MPGNN), and produces scalar embeddings representing features of nodes and edges of the graph and vector embeddings representing geometric relationships of the graph. The system processes the scalar embeddings via a vector scalar interactive message passing mechanism of a message passing sub-block of the MPGNN to generate and pass scalar information from the scalar embeddings to an embedding space containing the vector embeddings. The system updates the vector embeddings based on the embedding space containing the scalar information and the vector embeddings. The system updates the scalar embeddings based on run-time geometry calculations of the geometric relationships encoded in the vector embeddings. The system computes an updated molecular graph based on the updated scalar and vector embeddings and outputs a target molecular property value based on the updated molecular graph.
Inventors:
WANG TONG (US)
SHAO BIN (US)
LIU TIEYAN (US)
SHAO BIN (US)
LIU TIEYAN (US)
Application Number:
PCT/CN2022/126834
Publication Date:
April 25, 2024
Filing Date:
October 21, 2022
Export Citation:
Assignee:
MICROSOFT TECHNOLOGY LICENSING LLC (US)
WANG TONG (CN)
WANG TONG (CN)
International Classes:
G16C20/30; G06N3/02
Other References:
KRISTOF T SCHÜTT ET AL: "Equivariant message passing for the prediction of tensorial properties and molecular spectra", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 7 June 2021 (2021-06-07), XP081978697
PHILIPP THÖLKE ET AL: "TorchMD-NET: Equivariant Transformers for Neural Network based Molecular Potentials", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 5 February 2022 (2022-02-05), XP091151017
JOHANNES KLICPERA ET AL: "Directional Message Passing for Molecular Graphs", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 6 March 2020 (2020-03-06), XP081616091
WANG YUSONG ET AL: "ViSNet: a scalable and accurate geometric deep learning potential for molecular dynamics simulation", 29 October 2022 (2022-10-29), XP093035658, Retrieved from the Internet [retrieved on 20230329]
PHILIPP THÖLKE ET AL: "TorchMD-NET: Equivariant Transformers for Neural Network based Molecular Potentials", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 5 February 2022 (2022-02-05), XP091151017
JOHANNES KLICPERA ET AL: "Directional Message Passing for Molecular Graphs", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 6 March 2020 (2020-03-06), XP081616091
WANG YUSONG ET AL: "ViSNet: a scalable and accurate geometric deep learning potential for molecular dynamics simulation", 29 October 2022 (2022-10-29), XP093035658, Retrieved from the Internet
Attorney, Agent or Firm:
SHIHUI PARTNERS (CN)
Download PDF: