Skip to main content
Fig. 2 | Journal of Cheminformatics

Fig. 2

From: Prediction of blood–brain barrier and Caco-2 permeability through the Enalos Cloud Platform: combining contrastive learning and atom-attention message passing neural networks

Fig. 2

Diagram of the transformer-based MPNN. The framework consists of A a D-MPNN, B an atom attention multi-head transformer, and C a FFN to predict the property of interest (in this case the molecule’s ability to cross a biological barrier). Each component in this figure represents different vector representations within the model, as defined in the pseudocode in Table 1

Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us