Skip to main content
Fig. 2 | Journal of Cheminformatics

Fig. 2

From: Syn-MolOpt: a synthesis planning-driven molecular optimization method using data-derived functional reaction templates

Fig. 2

The overall workflow of Syn-MolOpt. A Model training stage; and B molecular optimization stage. During the model training stage, four neural networks were trained to predict the reaction action, the first reactant, the reaction template, and the second reactant. In the molecular optimization stage, these neural networks were used to predict the aforementioned four reaction elements respectively

Back to article page