Cargando…

Explaining compound activity predictions with a substructure-aware loss for graph neural networks

Explainable machine learning is increasingly used in drug discovery to help rationalize compound property predictions. Feature attribution techniques are popular choices to identify which molecular substructures are responsible for a predicted property change. However, established molecular feature...

Descripción completa

Detalles Bibliográficos
Autores principales: Amara, Kenza, Rodríguez-Pérez, Raquel, Jiménez-Luna, José
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Springer International Publishing 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10369817/
https://www.ncbi.nlm.nih.gov/pubmed/37491407
http://dx.doi.org/10.1186/s13321-023-00733-9