Your browser doesn't support javascript.
loading
MESPool: Molecular Edge Shrinkage Pooling for hierarchical molecular representation learning and property prediction.
Xu, Fanding; Yang, Zhiwei; Wang, Lizhuo; Meng, Deyu; Long, Jiangang.
Afiliación
  • Xu F; School of Life Science and Technology, Xi'an Jiaotong University, 710049 Shaanxi, China.
  • Yang Z; School of Physics, Xi'an Jiaotong University, 710049 Shaanxi, China.
  • Wang L; School of Life Science and Technology, Xi'an Jiaotong University, 710049 Shaanxi, China.
  • Meng D; Rearch Institute for Mathematics and Mathematical Technology, Xi'an Jiaotong University, 710049 Shaanxi, China.
  • Long J; School of Mathematics and Statistics, Henan University, 475004 Henan, China.
Brief Bioinform ; 25(1)2023 11 22.
Article en En | MEDLINE | ID: mdl-38048081
Identifying task-relevant structures is important for molecular property prediction. In a graph neural network (GNN), graph pooling can group nodes and hierarchically represent the molecular graph. However, previous pooling methods either drop out node information or lose the connection of the original graph; therefore, it is difficult to identify continuous subtructures. Importantly, they lacked interpretability on molecular graphs. To this end, we proposed a novel Molecular Edge Shrinkage Pooling (MESPool) method, which is based on edges (or chemical bonds). MESPool preserves crucial edges and shrinks others inside the functional groups and is able to search for key structures without breaking the original connection. We compared MESPool with various well-known pooling methods on different benchmarks and showed that MESPool outperforms the previous methods. Furthermore, we explained the rationality of MESPool on some datasets, including a COVID-19 drug dataset.
Asunto(s)
Palabras clave

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Asunto principal: Aprendizaje Profundo / COVID-19 Límite: Humans Idioma: En Revista: Brief Bioinform Asunto de la revista: BIOLOGIA / INFORMATICA MEDICA Año: 2023 Tipo del documento: Article País de afiliación: China Pais de publicación: Reino Unido

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Asunto principal: Aprendizaje Profundo / COVID-19 Límite: Humans Idioma: En Revista: Brief Bioinform Asunto de la revista: BIOLOGIA / INFORMATICA MEDICA Año: 2023 Tipo del documento: Article País de afiliación: China Pais de publicación: Reino Unido