doi:10.1063/5.0239722Nicholas CasettiPragnay NevatiaJunwu ChenPhilippe SchwallerConnor W. ColeyThe Journal of Chemical Physics
[LG] Molecular Hypergraph Neural Networks O网页链接 分子超图神经网络(MHNN)是一种针对有机半导体的光电性能预测的模型。MHNN通过引入分子超图来解决传统图模型无法准确表示多中心化合物键和共轭结构等高阶连接的问题。该模型在多个有机光伏相关数据集上表现出色,即使没有使用任何三维几何信息,也能超越基线模型的性能...
Note that higher-order networks and simplicial complexes can describe the many-body interactions beyond pairwise interactions. Hypergraphs are a further generalization of simplicial complexes. An hypergraph is composed of hyperedges, which are formed by a set of vertices. Recently, simplicial complexes ...
Hiroshi Kajino ICML 2019 grammar, VAE, hypergraph, goal-directed optimization GithubMulti-objective de novo drug design with conditional graph generative modelYibo Li, Liangren Zhang, Zhenming Liu Journal of Cheminformatics, 10 graph neural networks, distribution-learning, auto-regressive, conditional ...
representation. MathDL23combined differential geometry, algebraic graphs, and algebraic topology to form rotational and translational invariant molecular fingerprints. Based on persistent homology, Liuet al.24developed a persistent spectral hypergraph model, in which persistent attributes were used as finger...
Solutions to the handling of multi-valent bonds have been introduced via the use ofhypergraphs; in a hypergraph, edges are sets of at least two atoms (hyperedges) instead of tuples of atoms [37]. However, the use of hypergraphs is not further discussed here as they are not currently wide...
Liu and co-workers used persistent spectral hypergraph based molecular descriptors for their machine learning approaches to predict protein-ligand binding affinities. Vaisman and Masson have developed a computational geometry approach using Delaunay tessellations to capture the influence of the four nearest ...
In this work, we present FraHNN: Fragment-based Hypergraph Neural Network, a hypergraph learning framework that introduces molecular fragments to obtain more complete molecular information. Specifi-cally, FraHNN represents molecules as nodes of the hypergraph and fragments as hyperedges, using the ...