From bags of graphs to stereo subgraphs in order to predict molecule's properties

Pierre-Anthony Grenier (1, 2) , Luc Brun (1, 3) , D. Villemin (4, 3)

1 Equipe Image - Laboratoire GREYC - UMR6072
2 UNICAEN - Université de Caen Normandie
3 ENSICAEN - École Nationale Supérieure d'Ingénieurs de Caen
4 LCMT - Laboratoire de chimie moléculaire et thioorganique

Quantitative Structure Activity and Property Relationships (QSAR and QSPR), aim to predict properties of molecules thanks to computational techniques. In these fields, graphs provide a natural encoding of molecules. However some molecules may have a same graph but differ by the three dimensional orientation of their atoms in space. These molecules, called stereoisomers, may have different properties which cannot be correctly predicted using usual graph encodings. In a previous paper we proposed to encode the stereoisomerism property of each atom by a local subgraph. A kernel between bags of such subgraphs then provides a similarity measure incorporating stereoisomerism properties. However, such an approach does not take into account potential interactions between these subgrahs. We thus propose in this paper, a method to take these interactions into account hence providing a global point of view on molecules's stereoisomerism properties.

Keywords: Chemoinformatics; Stereoisomerism; Graph kernel

Domains: Chemical Sciences; Cheminformatics; Computer Science [cs]; Artificial Intelligence [cs.AI]; Computer Science [cs]; Computer Vision and Pattern Recognition [cs.CV]

Pierre-Anthony Grenier, Luc Brun, D. Villemin. From bags of graphs to stereo subgraphs in order to predict molecule's properties. International Workshop on Graph-Based Representations in Pattern Recognition, 2015, Beijing, China. pp.305-314, ⟨10.1007/978-3-319-18224-7_30⟩. ⟨hal-01848014⟩