Deep contrastive learning of molecular conformation for efficient property prediction

被引:0
|
作者
Yang Jeong Park
HyunGi Kim
Jeonghee Jo
Sungroh Yoon
机构
[1] Seoul National University,Department of Electrical and Computer Engineering
[2] Seoul National University,Institute of New Media and Communications
[3] Massachusetts Institute of Technology,Department of Nuclear Science and Engineering
[4] Seoul National University,Interdisciplinary Program in Artificial Intelligence
来源
关键词
D O I
暂无
中图分类号
学科分类号
摘要
Data-driven deep learning algorithms provide accurate prediction of high-level quantum-chemical molecular properties. However, their inputs must be constrained to the same quantum-chemical level of geometric relaxation as the training dataset, limiting their flexibility. Adopting alternative cost-effective conformation generative methods introduces domain-shift problems, deteriorating prediction accuracy. Here we propose a deep contrastive learning-based domain-adaptation method called Local Atomic environment Contrastive Learning (LACL). LACL learns to alleviate the disparities in distribution between the two geometric conformations by comparing different conformation-generation methods. We found that LACL forms a domain-agnostic latent space that encapsulates the semantics of an atom’s local atomic environment. LACL achieves quantum-chemical accuracy while circumventing the geometric relaxation bottleneck and could enable future application scenarios such as inverse molecular engineering and large-scale screening. Our approach is also generalizable from small organic molecules to long chains of biological and pharmacological molecules.
引用
收藏
页码:1015 / 1022
页数:7
相关论文
共 50 条
  • [1] Deep contrastive learning of molecular conformation for efficient property prediction
    Park, Yang Jeong
    Kim, Hyungi
    Jo, Jeonghee
    Yoon, Sungroh
    [J]. NATURE COMPUTATIONAL SCIENCE, 2023, 3 (12): : 1015 - +
  • [2] GeomGCL: Geometric Graph Contrastive Learning for Molecular Property Prediction
    Li, Shuangli
    Zhou, Jingbo
    Xu, Tong
    Dou, Dejing
    Xiong, Hui
    [J]. THIRTY-SIXTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FOURTH CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE / THE TWELVETH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2022, : 4541 - 4549
  • [3] Molecular property prediction by semantic-invariant contrastive learning
    Zhang, Ziqiao
    Xie, Ailin
    Guan, Jihong
    Zhou, Shuigeng
    [J]. BIOINFORMATICS, 2023, 39 (08)
  • [4] 3D graph contrastive learning for molecular property prediction
    Moon, Kisung
    Im, Hyeon-Jin
    Kwon, Sunyoung
    [J]. BIOINFORMATICS, 2023, 39 (06)
  • [5] Advanced deep learning methods for molecular property prediction
    Pang, Chao
    Tong, Henry H. Y.
    Wei, Leyi
    [J]. QUANTITATIVE BIOLOGY, 2023, 11 (04) : 395 - 404
  • [6] Pretraining deep learning molecular representations for property prediction
    Liu, Bowen
    Hu, Weihua
    Leskovec, Jure
    Liang, Percy
    Pande, Vijay
    [J]. ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, 2019, 258
  • [7] Deep learning methods for molecular representation and property prediction
    Li, Zhen
    Jiang, Mingjian
    Wang, Shuang
    Zhang, Shugang
    [J]. DRUG DISCOVERY TODAY, 2022, 27 (12)
  • [8] DGCL: dual-graph neural networks contrastive learning for molecular property prediction
    Jiang, Xiuyu
    Tan, Liqin
    Zou, Qingsong
    [J]. BRIEFINGS IN BIOINFORMATICS, 2024, 25 (06)
  • [9] Molecular property prediction by contrastive learning with attention-guided positive sample selection
    Wang, Jinxian
    Guan, Jihong
    Zhou, Shuigeng
    [J]. BIOINFORMATICS, 2023, 39 (05)
  • [10] Applications of Deep Learning in Molecule Generation and Molecular Property Prediction
    Walters, W. Patrick
    Barzilay, Regina
    [J]. ACCOUNTS OF CHEMICAL RESEARCH, 2021, 54 (02) : 263 - 270